Public Disclosure Authorized TBE STUDY OF TEE SUBSTITUTION OF LABOR AND Public E Q U I M N T IN CIVIL CONSTRUCTION (SOL) Disclosure Authorized A RESEARCR h IMPLEMENTATION PROJECT Public FOR R E C O N S ~ ~ ~ IAIJD DEVELOPMENT O N PRoJBCT COMPLETION REPORT Disclosure 0Ci 2 t 1991 Authorized Public Disclosure Operations Policy Staff Transportation Department Authorized - ~ . i ; h r a and tbaruraa 1 canciaacar (en) 0.3937 i n c h u (in) 1 -car (a) 39.37 lachas (in) o r 3.28 faat ( f r ) 1 eubic w e a r (a3) 35.31 eubic faat (cu f t ) 1 haccara (ha) -- 2.471 a a a s a 0 8 W of Abbrariaciosr Africu o..r10prnc In& *rid wnlr. nan cancral m a g r a n l l i a g hojacc ~ c b c i c a clu a r b a d coawcruccios by concraecora, w i n g labor-incanriva m c h d a (8ocrduru) -parerut of Labor-hcanaiva h a d t%~tn1eCion:B0nd~8# maericr D.ralop.nt buncila (Xanya) Dlraecoraca-C.nara1 of m a 1 h a & (maieo) Teadar had. Wvisioa ( h a i n ) h m c s CIA Taadar b a d Colucruction nnic (mnduraa) l i a u l iaar (7111-6/30) CH: Graulliug, Bridge d Culvarca R o w a m 8 (Kaaya) Qn? croaa D a a r i c ~ o d u c c 001 Ooramwnc of h a i n OOC b n r o r n c ot Qud OODI Ooranhnc ot cha' D d 8 i c f . n h p u b l i c 00a b u r n m n c .of Boaduras at Ooramrnc ot may. 00L b u r a r a c ot t a o c h o nu) I n r a r ~ c i o l u lBank for &eoo.cruccioa d D.r~loplwnc, World h a k IDA h c a r a ~ c i o a r lDaulopmnc haociacion ID8 Incar-Amrican D8ralop.nC 98nk IU) I n ~ a r ~ ~ c i oLabor Office t u l KZCP h a y 8 &at Crav*llin( Rojact tC(I Lbor-Incanriva bwcruccion h i t (Laoeho) n Nsloci (hsotho) m Mnircry of Transpare aad Col.unicacion8 (buy.) N.A. Ibc rr.ilabla I b m g i a n M e n q for D.ralop.ac EricLh Oraraaaa Davatopmc kidniacr8cion h r q r a p b h a u r c b and kpl.uncacion ~ o j a c I.... t Tha Scud, on cha k b r i r u c i o n of Labor 8nd i q u i p . 8 ~i n Civil b ~ c r u c c i o n a Lrd (MA) w b r a 1 Aacua h a & Rogrpl ( h a m ) . me l u o a 01 cha DaLnician bpublic IWA mpublic 01 Souch Mrica SIDA Suadiah hcam8ciorul Daralopmnc Mane, SOL m a Subcitueion of t b o r and tquipmoc i n Civil brucruccioa ( m i * acronym h. k a n luad coaaircancl7 t o idantif7 cha study u a whola) hanrportacion Dapartmac of cha World 98ak Briciah haluporc and Road kraarch laboratory Tachnolow PDic Micad N8eioua D.v~lapmncRosr- Uuiemd Kingdom h i c a d Sc8c8a Aqancy for Incarnational D8veLopwnc USS UnLCrd Seacar Dollars TEE STUDY OF TEE' SUBSTITUTION OF LABOB'AND EQUIPMENT I N CIVIL CONSTRUCTION A RESEARCR AND TMPLENENTATION PROJECT PROJECT C~MPLETIONREPORT Basic Data Sheet................................................. i v Summary and Conclusioas.......................................... v - REPORT 1. INTRODUCTION................................................. 1 2. PUBPOSE OF TEE PBOJECT....................................... 2 3.1 Reparatory Stage.................,..................... 2 3.2 Research Stage....................................... 2 3.3 Field Testing Stage..................................... 3 4. OPERATIONAL TESTING OF.TBE RESEARCH FINDINGS................. 5 4.1 Honduras................................................ 5 4.2 Konya................................................... 10 4.3 Bank-sponsored Rograms fn other Countries.............. 17 4.4 Rograms-without Bank Support........................... 25 5 . FINANCIAL EVALUATION OF THE CONCEPT, AND ECONOMIC ASPECTS.... 28 I. EONDURBS: Variation of Costs per Kilometer f o r a Typical Road i n Flat Terrain as a Function of the Volume of Earthwork. 2. HONDURAS: Cost Comparison between Work Methods - Procurement Arrangements for Cases Investigated. 3. HONDURAS: Costs per Kilometer of Road i n Flat Terrain as a Function of Unit Prices and Quantities for each Group of Projects and a Typical Road Section. EONDUlUS: Costs br Uonmter of m d in Mountainous Terrain as a Function of U n i t Rices and Quantities for each Group of Projects and a Typical Road Section. HoNDmAS: Costs per Kilometer of a Typical b a d in Flat Terrain acl: a Function of an krcreaaed/D.creued Labor or Equipmant Componen:. HONDURAS: Smmury of Findings Related to Appropriate Technologies. - Percentage -9- HONDURAS: FEEDEB ROAD PROGRAM Distribution of Cost Categories. KENYA: Cost Comparison between the Rural Access Roads Program and - the two Gravelling, Bridge and Culverts Rajects - Average for 1981182 1983184. BENIN: Typical Mstribution of Major Equipment to Brigades. - Number BENIN: Local Employees and Paymentsfor 1980. INDIA: Comparat.iqe Unit Costs '.of Urtliworka by Different Methods as a Function of Haul Distance and 'Wage Bate. KENYA: - Basic Rural Access Boads Rogram Organizational Structure. . KENYA: Rural Access Roads Rogram - Unit and Site Organization Cliart MEXICO: Organization Chart for a Directorate-General of Rural Roads - Establishment in a State. - . Screening at CountrylRegion Sector Levels. Screening at RogramIRoject hvels. Suitability of Rograma/Projects for Construction by Labor-Based Matho& a t Different Unskilled Labor Wages. - ANNEX Project-Related Publications: 1971-1985 PREFACE 1. In 1971, the World Bank launched the "Study of the Substitution of Labor and Equipment i n C i v i l Construction" (SOL) an a framework for research on, and demowtration of cowtruction =thocia appropriate for, the s o c g e economic e n v i r o m n t of laborabundant and capital-scarce countries. In the -.z- f a l l of 1973, a group of industralized Bank member countries namaly Canada, Denmark, Finland, the Federal Republic of Germany, Japan, Norway, Sweden, the M t e d Kingdom, and the Dnited States, joined the Bank i n establishing a Trust Fund to support the SOL activities. 2. W e have aimed to provide, i n this Project Completion Beport, a compen- d i m of the most important results achieved by the SOL Roject i n terms of research and practical application of the findings. A short assessment of the i n i t i a l project implementation, i.e. preparatory, research, and f i e l d testing stages, is followed by a more detailed account of the operational testing of the findings during the implemantation phase, complemented with observations from other countries where there were no direct SOL interven- t i o n ~ . We hope that the presentation of the problems, solutions, and les- sons of the practical application of the SOL experience may be of particular assistance for the dissemination of labor-intensive technologies i n the future. 3. Individuals of many nationalities, professions, and backgrounds parti- cipated i n t h i s SOL Roject, including: staff members of the World Bank and particularly of its %ansportation Department (TRP); other international organizations, bilateral agencies and consulting firms; government officials and representatives from private and co-nity organizations; large and small contractors; and thousands of laborers i n Africa, Asia and Latin America. Among a l l participants special mention must be made of C e l l G. Harral (TRP). who initiated the SOL Roject, directed it during its research and early implementation phases, and l a t e r supported it with constructive suggestionm and encouragement. 4. +inloas and value judgements are those of the team that carried out the SOL Project and do not necessarily reflect the official position eithar of the World Bank o r of any of the governments that assisted i n the study, financially or otharwise. TEE STUDY OF TEE SUBSTITUTION OF LBBOR AND EQUIPMENT I N CIVIL CONSTRUCTION A RESEbBCE AND IMPLEKENTATION PROJECT PROJECT COMPLETION REPORT BASIC DATA SEEET g a y - ~ o j e Data c ~ -= Actual cL Gtimate lJorld -- Others Total - Bank Total Project Cost (US$million) : Phase I and 11 Phase I11 Total Grant Phases I I1 Total Project -Completion Dates 10171 10173 06/86 Cmulative Actual Msbursments a/ - (US$ thousands) MRLD BANK' Phase I & I1 Phase I11 Others: Phase I11 Grand Total WORLD BANK Phase I & I1 Phase 111 Others: Phase 111 Grand Total -a/Note: FY74 through 86 are related to the annual audit report periods (from March 1 t o end of February f o r the respective years). SUMWRY CONCLUSIONS 1. By the mid-1960's many people were beginning t o question the appropriateness of transferring the technologies of industrial countries to the Third World, without adaptation t o the different factor endowments and circumstances. A t the same t i m e ,the improvement of the standard of living i n rural areas through the upgrading of infrastructure, the generation of employment for the unskilled rural poor, and the conservation of scarce - foreign exchange were included m u g the objectives of policy makers i n many developing countries. Yet, there was a general reluctance to employ labor- intensive techniques f o r c i v i l construction. No systematic investigation had been carried out on the use of large numbers of otherwise un- or under- employed wrkers t o build civil engineering works. This moved the Bank to launch i n 1971, this Research and Implementation Project under the t i t l e of the Study of the Substitution of Labor and Equipment i n C i v i l Construction, or SOL for short. 2. The Project was implemenated i n several stages. Research work i n the f i r s t half of the 1970's showed that under certain circumstances labor- intensive construction could compete with machines on both technical and financial terms. Furthermore, the competitiveness of labor-based methods could be improved through better organization and management, incentive paymant systems, better tools and auxiliary equipment, and interventions with food and mdicines to upgrade the health and nutrition of the workers. Up t o that point, investigations of the technical and econodc feasibility of labor-intensive methods had focussed on Asian countries, particularly India and Indonesia. Indications of success i n Asia suggested that experimentation should expand to other developing countries. Beginning in 1975, the World Bank research team became involved i n the design and execution of labor- intensive programs i n Eonduras, Kenya, Chad, Benin. Lesotho and the Dominican Republic. Many of the lessons from the earlier work i n India and'Indonesia were tested, applied and refined. From observation and research, the em- phasis shifted to the planning and execution of expanding labor-intensive operations carried out by a central organization at a large number of-sites. Other developing countries also benefitted from the Project findings, and these were complemented by the experience gained i n countries, notably Mexico, where labor-intensive work program were carried out without the direct involvement of the Bank. 3. Fixperfence showed that major labor-intensive contruction operations could be mounted i f serious efforcs were made and an abundant supply of w e l l - managed labor wan available a t competitive wages. In the mid-197O1s, those wages were found t o l i e belaw a threshold of $2 per day for most labor- intensive operations. Since then the wages threshold has increased to some $4 per day, depending on the productivity of the laborers and the nature of the work. It is therefore concluded that the application of labor-intensive techniques should merit close attention for many types of work i n some 50 countries where rural wages are the equivalent of about $4 per day or less. 4. h s p i t e the increase of the threshold i n current dollars, wage rates u e not significantly higher, i n real term, than those of the mid 1970's. If anything, capital is more scarce now and, because rural unemploymant has not decreased i n many countries, labor continues to be abundant, making the labor-intensive option even more attractive. '1 'However, for countries with no traditional experiencs with labor-Tntensive technologies, it was found that a dnimum of three years of preparation is -- required for mobilization, staff training and the introduction of specialised institutional arrangemants, before the large-acale me of unskilled labor can start. --=. 5. , In mnm~ry,the principal lessons learnt from the Roject were: (a) Labor-intensive work methods are techpically and finandally feasible for certain types of work under certain conditions. (b) There is a significant potential for the improvement of the efficiency of existing labor-intensive work methods where they are traditionally .used. (c) IO. countries with minimum wage levels of up to US$Z.OO per day, suitable types of civil works should be exe- cuted by labor-inte~ivework mathods, uulass there are apecifie reasons against such a concept (8.8.. due to the non-availability of the required labor force i n the project area). mere the minimum wage ranbes between $2.00 and $4.00 per day, the use of unskilled labor should be conscientiously coasidered; and with minimum wages slightly above $4 per day, the appropriate mix of labor and equipment may s t i l l tend towards labor- -intensive use under certain circumstances (e.g., in- volving small quantities i n remote areas). (d) The strong and sustained commitment of the government to the cost-effective use of unskilled labor and adequate i ~ t i t u t i o a a lsuppore are prerequisites for a successful executioo of such projects or programs, in countries which traditionally have not used labor-fntensive .. methods. e Labor-intensive works are also management-intensive, and require special attention i n this respect, especially where labor relations are not vell developed. This points to the special faportance of certain cadres of -11 - The World Bank, "The Construction Industry Issues and Strategies in De~lopingCountries" (IBRD, Washington, LC., 1984) p.43. capital requirermnts ard low. W l e a11 the projects referred to i n this report (excepting the uses of India) b.o. been done by force account, payments to -ricers are m i c a l l y based on production, hence the execution lends itself well to contract work. In spite of this, in most countries, appropriate labor-intensive technologies have failed to interest private sector contractors. A major reason for this has k e n price distortions - while governnrnt may pay labor by its economic value, private industry is often bound by miaimurn wage laws and labor union agreements, which raise the cost of hiring uaakilled labor substantially. Bcnce, t o encourage the progress of labor-intensive worka contracting, which we see a#an hportant element i n setting up suitable programs, the question of wage rates needs t o L bn resolved i n the cases concerned so that the financial costs faced by the contractor reflect the econondc costs. 7. The studies covered i n this report have shown that an early invol- vement of the communities to bs served by the completed works is essential: their collaboration mst be obtained for the supply of labor and the provi- sion of rights-of-way, easements, and quarries for materials. But beyond chis, there is a need in many countries or areas, for community or regional initiatives to develop imaginative ways of achieving a better balance in factor use i n construction. Examples of such initiatives include comeunity and individual self-help construction schemes, community work programs com- bining wage payment wtth some recourse to self-help principles to reduce the effective average price of labor, and other demonstration programs on the use of efficient labor-intensive techniques i n construction. 8. The governments of developing countries w i l l continue to need external funding for a long time to come, to establish, improve, or extend labor-intensive work programs. Likewise, it w i l l be necessary to assist them ' with technical expertise. In order to enhance the resources available for evaluation and supervision of labor-intensive schemes, the World Bank is closely cooperating with the United Nations Development Programme and the International Labor Office. Elaking use of the findings and~conclusionsof this Research and Implementation Project, joint activities are being carried out, or are planned to be executed within FP1986, i n 19 countries of which 12 are located in Africa, south of the Sahara. This Project has generated an impressive multiplier effect i n the awareness of people i n developing as well as i n industrial countries, that labor-intensive work methods offer a viable alternative, under certain conditions, to conventional, i.e., equipment- intensive, techniques and often provide a more efficient use of domestic resources. THE STUDY OF THE SUBSTITUTION OF LABOR AND EQUIPbBNT I N CIVIL CONSTRUCTION A RESEARCH AND IMPLEMENTATION PROJECT PROJECT COMPLETION REPORT 1.1 R i o r t o the 1960's the appropriateness of the technology used f o rs- c i v i l work8 construction i n specific countries was not much of a concern i n the developing world. With the exception of a few Asian countries, most developing countries, with the f u l l support of the World Bank (the Bad) and other aid agencies, w r e bent on as rapid a mechanization as was possible. This largely meant transferring the capital-intensive standards developed in the high-rsage, capital-abundant countries of Europe. North Amrica and Asia t o the low-income, capital-scarce developing countries without questioning whether these technologfes were appropriate. For example, the dominant consideration for the formulation of highway design standards i n the industrial countries was t o increase travel speed and relieve congestion because of the high value attached to travel time savings of passenger t r a f f i c i n high-incoma countries. But i n developing countries, the prime consideration is often a question of access, i.e., to get farm products to the markets, rather than. of speed. 1.2 By the mid 1960's many people, both wlthin and outside the Bad, were beginning to question the appropriateness of such wholesale transfer of technologies without adaptation t o the different factor endorrments and circumstances of developing countries. However, a general reluctance i n employing labor-intensive techniques for c i v i l construction still existed. This reluctance was caused by the problems thought to be inherent in the traditional labor-intensive methods, including for example: (a) fears of poor quality of the finished work, (b) fears of slow progress i n executing the construction work, (c) concern about the need t o make special organizational . arrangements, and (d) concern that the costs would be higher than those of equimnt-intensive operations. 1.3 Following a major procuremant dispute with a member country i n 1969, the Bank's management announced a program of research on highway construction technologies on August 18. 1970. As a direct result of this decision, the study on the Substitution of Labor and Equipment in Civil Construction (SOL, or the Project) was launched i n March 1971 as a research and implementation project. 2.1 The purpose of the Roject was to establish the technical feasibi- lity,of alternative c i v i l construction technologies utilizing different combinations of labor and equipment and to relate those technically feasible technologies to relative factor scarcities so that the economically optimum technology could be detemdned i n any given case.l/ 3. m-ATION OF TBE PROJECT 3.1.1 Phase I of the Roject (February-October 1971) focused on a survey of existing literature and problem formulation for road construction. Specia- lized staff from four consulting f i r m (from the United Kingdom and France) were employed to form a team of engineering and economic consultants. They were to assess whether and to what extent it was technically and economically feasible to substitute labor for equipnuant i n road construction. h e find- ings of the Phase I report (Annex, para. 11, were that tha substitution is technically feasible for a wide range of construction activities for roads of various qualities. The economic feasibility depends oa relative factor pricea and factur productivities under different conditions. Hwever, the engineering data wrmally available were found t o be i ~ p p r o p r i a t e l ystruc- tured or too poorly defined with respect t o specific enulrommntal conditions .to pandt a quantitative analysis of the substitution possibilitfes. 3.2 Easearch Stage 3.2.1 Phase 11. which began in November 1971 and finished in October 1973, focussed on direct field observations of ongoing construction activities to obtain the production relationships between varying inputs of equipment and labor, and the output of different tasks of civil construction. Field data were collected on 30 road, dam and irrigation construction sites. In Indonesia, studies were also conducted to determine the effect of health and nutritional standards on the productive potential of the labor force. 3.2.2 - Aa observed in the course of the study, labor-intensive techniques, as traditionally practiced for road, hydro-power and irrigation mrks i n India and Indonesia. were not competitive with modern equipmaat-based methods a t 1972 prices. This was because (a) primry emphasis was on employmant creation rather than efficient w e of labor, and/or (b) the existing methods -re primitive and did not employ elemerrtal mechanics effectively, i f a t all. As a consequence, labor productivities were extremely low. -21 World Bank, "Substitution of Labor and Equipment in Civil Construction - A Study Proposal and Progress of Work" (Washington, D.C. September 1974). 3.2.3 It was, however, noted that there are important margins for labor productivity increases through (a) improved organization and management, (b) improved tools and hardware, and (c) upgrading of the health and nutritional status of the workers. The findings of Phase 11were summarized i n the "Final Faport" dated January 1974 (see para. 1 of Annex). 3.2.4 A retrospective critique of the above Faport suggests that estimated costs for ope ratio^ using the modern equipmnt-intensive methods -- were based on rather optimistic assumptions related t o equipmant utilization, -?- and comequently resulted i n costs which -re too low t o be representative of the working conditiom in most developing countries. To study the effect of different parameters of work execution i n Phme 11, the various components of civil work c o ~ t r u c t i o nwere broken into the wst minute elements of tasks and activities. In excavation work, for instance, the factors studied included s o i l types, moisture content and ambient temperatures; loading was identified as a separate activity, with loading height as a major parameter, and so on. The amount of disaggregation could be questioned. However, the disaggregation was useful, as a means for identifying the key parameters i n the resources appropriate for a particular site or project. 3.3 Field Testing Stage 3.3.1 A8 a result of the findings i n Phases I and 11, a decision was made t o continue into Phase 111of the Project. The original scope of work for P b e 111was implemented from September 1973 t o August 1976, comprising work commenced i n India i n October 197-3 and i n Indonesia i n September 1974, followed by Honduras i n September 1975 and Kenya i n November 1975. This was followed by further practical application cases, completed i n September 1985. The objective of this third Phase was to develop and demonstrate coet- effective labor-intensive technologies i n a number of different countries encompassing different environmental conditions. An economic solution for the pressing problem of creating productive emplopnt opportunities for large segments of unemployed and underemployed population i n many developins countries was t o be sought. The necessary improvements i n labor productlvicv were to be achieved mainly by: a improvements i n the organization and management of the program aa a whole, and of construction sites; (b) introduction of improved tools and equipment specifically designed or adapted for we by casual labor; and (c) improvements i n the health and nutritional condition of the labor force 3.3.2 The following Project activities were carried out i n India: - (a) Experimental Studies At four different sites related to: ( i ) stone crushing by manual and mechanical methods. ( i i ) testing and use of different designs of wheelbarrows, (iii) use of light-railway wagons for haulage of boulders, (iv) use of a pulley atrangemant for hauling and for a l i f t of 18 mters, (v) reorgaaization of excavation along hillsides, and - (vi) experiments with hauling earth with tractortrailers, wheel-barrows, bullock carts and donkeys. Increases in productidty ranging from 10 to 200 percent were achieved. - (b) PirstDeaonstration Project Brcavation on hillaides, and --. road surfacing with a.2.5 cm premixed asphalt macadam carpet. - Tha (c) Secd Demodtration Roject a w t r u c t i o n of a new road, including earthworks for a bridge approach and pavewnt coa8truction. . - (d) Health/Nutrition Study Several deficiencies were identified. The administration of iron raised worker output by 25 percent. - (a) Canal Construction The irrigation authorities made available a section of major canal construction for Roject experimenta- tion and observations. The productivity achieved i n major earthworks using existing methods (including the use of truclcs, camels, donkeys and headbaskets) were measured. Expe- r i ~ n t were also carried out with a monorail, railcart, power s winch and various types of vheelbarrm for earth moving, and studies conducted on hand tools and the effects of temperature. 3.3.3 In Indonesia, expertmental studies on earthwork- tasks were carried out on the construction of a 200 m length of irrigation canal and included: (a) excavation using ox-drawn ploughs and scrapers; (b> haulage using lever cranes, ropeways, ox-drawn scrapers, oxdrawn carts, and vheelbarrows; and (c) compaction with animaldrawn rollers and manual tampers. 3.3.4 In Honduraa and Kenya, the cooperation between the Governmant and the Bank was based on Memoranda of Understanding indicating that the parties would jointly pursue experimental and research activities intended to promote the development and implementation of labor-intensive technologies for low- class road construction and maintenance. The resulting activities i n Honduras and Kenya are reviewed in detail in Chapter 4 of this report, together with other pilot projects. 3.3.5 The i n i t i a l findings of Phase TI1 were summarized i n reports and technical memoranda (see l i s t in Annex). The experimental work on toad and irrigation sites partially dispelled the atmosphere of doubt about the possibility of improving the efficiency of labor-intensive construction methods, f i r s t reported in the Phase I1 findings. For certain construction tasks, the increase i n labor productivity realized from improvements i n basic tools, worker incentives, and site management made labor-intensive methods competitive with equipment a t base wages up t o about USS1.50 per day at 1976 price8 (Figure 1). With wage incentives and better quality shovels, f o r . example, labor productivity in excavation of hard s o i l s was observed to r i s e 150 percent (Annex, Technical Memorandum No. 2). After the introduction of improved procedures a t the site level, it WM concluded by about mid-1975 that, if labor-intensive construction methods were t o bo competitive at a large scale with equipment-based mthods, improved procedures for planning, -- financing, staff training and progress monitoring had t o be implemented a t the a r l t i - a i t e , program level. 4. OPEBILTIONAL TESTING OF TEE RESKARCB FINDINGS 4.1 Honduras 4.1.1 With growing evidence on the competitiveness of labor-intensive methods under pilot-site conditions, attention shifted towards the design and execution of m l t i - s i t e programs. In February 1975 a member of the Bank research team was invited by the Central American Bank for Economic Integra- tion t o give a seminar i n Tegucigalpa. Honduras, on the findings of the Project. Following the s e n i ~ m r ,the Bonduran Minister of Comnunicatione, Public Worlcs and Transport expressed interest in having the Bank study team contribute t o a nationwide program of construction of feeder roads, financed by t h s Bank's Sixth Highway R o j e c t and an agricultural development loan from ihe United States Agency for International Development (USAID). A Memorandum of Understanding stipulating the cooperation between the Government of Bonduras (GOB) and the Bank was signed i n September 1975. 4.1.2 Previous efforts made-by GOB t o use unskilled labor for construc- tion work after natural catastrophes had failed because no previous planning was made and very l i t t l e technical guidance or supervision was provided, resulting i n low labor productivity and poor construction quality. In the i n i t i a l stage of the new pilot project, emphasis was therefore placed on demonstrating the technical f e a s i b i l i t y of labor-intensive construction of low-traffic rural roads on selected sites. To support such activities, a two-year, US$400,000 program of technical assistance was established, financed i n equal parts by the GO8 (out of the Bank highways loan) and a grant from the SOL R o j e c t budget. -0 engineers of a German consulting firm started work i n April 1976, together with three Honduran engineers and one economist. The Feeder Road Construction Unit (FRCU) was established i n mid-1976 and comprised a work force of about 140 laborers. 'he positive results of the d e m o ~ t r a t i o nexercise, as w e l l as the positive response from the people i n the work area, led the GOB t o steadily increase the budget f o r labor-intensive construction i n the following years. 4.1.3 In January 1978, FRCU was upgraded, and became the Department of Labor-intensive Road Constructioa (Departamento de Caminos por Mano de Obra, DCMO) under the General Directorate of Highways. The technical assistance provided by the German consulting firm was extended u n t i l September 1980, a t which time DCMO became solely staffed by Boaduran (or Central American) personael who competently continued the planning and implementation work. By the end of 1982, LXMO employed 15 engineers, 1 econodst, other technical and administrative s t a f f , about 70 construction foremen, and over 2.000 laborers. The funding i n the 1976-80 period waa provided ( i n addition t o GOB budget allocatioas and loan proceeds) by grants from the SOL R o j e c t , the World Food Program, Gorman Food Aid. and USAID. DCMO's affidency aqd oatisfactorg parformrace later also attracted external a s s i s t a n ~ efrom Swiss -= - Aid, the Inter-keriun Development Bank (IDB), Kreditanstalt f u r Wieder- aufbau and the Europeaa Economic Commission. The B8nk'a Seventh Bighway Project included a component f o r continuing financial support of DCHO activi- t i e s The support given t o DCMO by the SOL and Bank project components was soon followed by a coasiderable influx of assistance from other international agencies, and steady funding from the Goveramnt's budget, t o the point that, i n the recently completed Eighth Highway R o j e c t the Bank's participation was circumscribed t o a monitoring brief, incorporated i n the loan agreement through a supplemental letter. 4.1.4 DCMO constructs or improves rural access roads f o r llght t r a f f i c (vehicles up t o 5-ton payload) with a 10 t o 15 cm thick wearing course of gravel over a width of 4.5 to 5 m. During 1976-80, the output amounted to 296 km; i n 1981/82 about 200 lem ware added; and by the end of 1984 a t o t a l of about 700 lem of rural access roads had bean coastructed. DCMO has established its own -intenance division, employing workers who had parti- cipated in coastructing the roads. For a few road sections, villagers have organized maintenance by communal unpaid work. These arrangements, although i n i t i a l l y coaaidered.to be temporary pending takeover of the roads by the Departneat of Maintenance, may i n f a c t need t o be established f o r long-tern maintenance, i n view of the budgetary constaints which r e s t r i c t the acti- v i t i e s of the Department of Mainfenance t o the priority network. 4.1.5 The principal resource for execution of the work, unskilled labor. w a s readily available i n the program area. The low percentage of absentaeiss indicated that there were few alternative employmnt opportunities f o r ua- skllled workers i n that area, apart from agriculture, and the program provided, on average, employment f o r about 1%percent of the economScall~ active male population of the program area. DCHO recruited the laborers a f t e r contacting the conrm'nity leaders of the villages to be served by each road. Skilled labor and supervisory staff were largely hired i n nearby com~unities. The payment procedure for unskilled labor was lengthy since .n ministries and several departments were involved requiring 35 administrat:** stepo. I n i t i a l l y , workers were paid by check a f t e r 8 to 15 weeks. This improved l a t e r , but payments coatiwed to be effected by check. Becausa 21 the payment by check i n areas where there were no banks, a market developed for discounting pay checks "on the spot," for a discount amounting on aver a(* t o the equivalent of between 1 and 2 days' work. On a more recent, Swiss- financed rural road project, a revolving fund was established to facilitaca cash payments, and efforts are being made t o use a similar f a c i l i t y on al: Dm0 projects. The early problems of delays i n cash payments were dtigaccd to soma extent by providing partial payment i n food. One donor provided food conforming t o the national diet, which waa readily accepted by the workers. The food supplied by another agency was, however, foreign t o Hondurans who sold it a t a fraction of its value t o merchants from outside the area. "Food for work' continues t o be the principal form of payment for maintenance work carried out by mo. 4.1.6 An institutional constraint affected the recruitment of profes- sional staff i n the early days of DCMO. 'Ihe central recruiting agency of COB required that all technical Lay positions should be f i l l e d by candidates -;= holding a university enginnering degree. To overcomc the shortage of appli- cants and, consequently, the heavy dependence on expatriatea. 608 agreed t o recruit aeaior undergraduates. From that point onwards, the engineering staff grew from three engineers (two of them expatriates) i n 1976 t o fifteen (a11 Honduran) by the end of 1982. This number dropped slightly afterwards, but it still includes two women, who share with their colleagues the f u l l range of field and office duties. 4.1.7 The cost of constructing or improving rural access roads i n Bonduras was analyzed a t various stages of program execution. A crucial as- pect of the cost evaluation was the leva1 of daily wages for unskilled labor, which was relatively high at an equivalent USS2.00 for construction mrkers from 1977 t o March 1981. Wages for construction workers were increased a t that time t o USS3.00 equivalent per day. Rowever, this cost increase was compensated t o a large extent by an increase i n productivity a~ workers strove t o retain their jobs, and the overall cost per ldlomater of coastruc- tion did not increaae proportionally. Another significant component of overall cost i n the early (or pilot) stage was that of the technical assis- tance by expatriates. This component amounted t o about 53 percent of overall costs for the i n i t i a l year (1976). decreasing to about 15 percent i n 1980. 4.1.8 The consultants carried out an exercise, comparing overall costs of building rural access roads by labor-intensive and equipment-based methods. This showed that the direct cost o f t h e 150 km ofroads built betveen 1976 and 1978 using labor-intensive methods amounted t o about ~ ~ $ 1 0 , 0 0per km, about half the amount for the conventional, equipment- 0 intensive methods based on tenders received by GOB. 4.1.9 To analyze. i n d e t a i l the more recent cost structure of constructing or improving rural access roads, the Bank commissioned a Guatemalan consul- tant t o coupare actual costs of labor- and equipment-intensive conatruction work executed under force account and contracts i n Bonduras. The report of September 1985 has provideda detailed review of the costs of various feeder road projects including those carried out by.DCMO (labor-intensive, force account); funded by the Bank and USAID (equipment-intensive, partly force account and partly under contract); and assisted by the IDB (equipment- intensive, a l l under contract). Furthermore, a hypothetical case (CMOCONT) was added, for labor-intensive construction under contract, using the basic data of DCMO performances, but wtth a 25 percent higher labor productivity due to an assumed better labor management. Four principal work categories were coated, namely: bush clearance (in hectares), earthworks ( i n cubic meters), drainage (in ilnaar meters), and gravelling (in cubic meters). The unit prices for each category ware analysed into their direct costs (including labor, equipment, materials, and tools); contractor's indirect costs (including insurance, financing charges, margin for profit/risk, and - taxes), and the indirect costs related to the executing or supervising agency (including supervision, administration, and consultant services). Equipment costs vmre further detailed indicating depreciation, interest on capital, maintenance, consumed supplias, and the wages and indfrect payroll expenses ----. of the operators. The assessment of these costs followed the qccepted prin- ciples of US cost analysis ntho&ology, and the results were confirpad by a c t d cammsrcial rental rates. Another Important input uaa the total average payroll cost for makilled laborers, irhich, after allowing for the cost incurred under prevailing labor laws, amounted to USS4.65 equivalent, starting from a minimmu wage of Lempiras 6.00 (USS3.00). 4.1.10 The analysis carried out by the consultant revealed the effect of varying volunes of certain work components (particularly the volume of earthworks) on the cost of construction by alternative methods. Assuming that OM typical kilometer of road i n f l a t or rolling terrain would require 0.9 ha bush clearance, 30 m of draina l and 870m3 of gravelling, the quantity 15 of u thworb f WM varied from 1,600 m /km to 5,000 03/km in steps of 200 m /km. Table-1 indicates the result of this assessment which c l e ~ r l y demonstrates that, i n spite of the relatively high labor costs, the labor- intensive method would have been the least expensive aolution for projects with less than about 2,300 m3/km of excavation. In mountainous terrain, the existence of rock formations in varying proportion to the total voluma.of earthwork may not provide adequate compatibility for a meaningful coat comparison. 4.1.11 The validity of this cost comparison is underlined by i t s being based on final costing information from actual completed work, encompassing a wide spread of procurement arrangements. Table 2 lists the arrangements considered. They range from force account work to comprehensive contracts, and include tw different contract/force account combinations i n between. Differences between force account and contract costs are identified, e.g., the cost of a man-day of casual labor is about 11.6 percent higher for a contractor than for a government agency, due to the different incidence of social laws upon each type of employer. This fact contributes to the higher costs for (XOCONT compared with DCMO as indicated i n Table 1. 4.1.12 Tables 3 and 4 summarize the cost ,comparison for f l a t and for mountainous terrain conditions, listing the principal work categories for the different project groups. B X O shows the lowest prices in f l a t terrain for its projects and for a typical road section with earthworks of about 2,000 m 3/km or below. For larger quantities of earthworks, the equipment-int ensive operation of USAID projects gave the least costs. The correlation of appro- priate technology to quantity of work is also confirmed for mountainous terrain. In this latter case, the equipment-intensive method for the project group with the largest earthworks (i.e., IBRD projects) proves to be the lowest, and this is confirmed for the typical road section with 4,500 m3/km. To t e s t the sensitivity of these results, Table 5 indieates the variations i n overall costs resulting from fluctuating labor or equipment inputs by in- creasing these costs to a level of 125 and 150 percent respectively, or de- creasing them t o 75 and 50 percent respectively. me point of balance (posi- tions "C" and "Ha)has been dram from the typical road section i n f l a t terrain of Table 3. 4.1.13 Zlm results of these sensitivity tests show the impact of the - varioua cost inputs and the effect 2- 011 the threshold level for different Cc intensities of labor and equipment uae. For prevailing conditions i n Bonduras, particularly a t a cost par man-day for &killed labor of USS4.65, the consultant's srrmmarized findings related t o appropriate technologies are shown i n Table 6. He established thresholds for bush clearance and earth- works; the l a t t e r is also used as a general indication for the other prin- cipal work categories i.e., drainage and gravelling. Even within the work categories, the assessment differentiates among activities, as e.g., he recommends to carry out excavation up to 2,500 m3/km by labor and above 2,500 m 3 / h by equipment while building ditches by labor, and levelling by equipment in both cases. This study also confirmed that the threshold for labor-intensive c i v i l works construction (at 1985 prices) l i e s slightly above a daily Age of USS4.00 equivalent: a t USS4.65, and under the conditions prevailing i n Bondurss, labor-intensive comatruction had become too expen- sive, 'except for roads involving minor volumes of earthmoving. 4.1.14 In conclusion, the positive results achieved with the.pilot project i n Honduras can be attributed to: (a) the COB'S commitment and positive attitude towards -labor- . intensive construction technology; (b) the establishment of a special Department for Labor-Intensive Road Construction within the permanent organization of the relevant millistry, providing both career opportunities to professional staff and a formal support structure for f i e l d operations ; (c) the flexibility i n overcoming some of the administrative and . procedural.constraints, particularly for the recruitment of staff; (dl the wst-effectiveness of the construction of rural access roads i n South-Western Bonduras; (e) a dedicated and effective program management, particularly a t the most vulnerable stage, during the i n i t i a l period; ( f ) a relatively f a s t replacement of the European expatriates who started the program by Hondurans; and (g) a source of c8sh incorns for poor segments of the populatioa: about 70 percent of the total expenditures for rural access road construction were disbursed aa wages to local labor (Table 7). This was particularly significant i n the southern . region, where about 80 percent of the labor force has no perwanant employmnt. 4.1.15 Purther improve.nts ham k e n recommended to the GOB, following &*- tho annual waitoring meetings set up under the Eighth Highway Project. Thuo include: (a) the need to pay laborers promptly i n cash; (b) the need to introduce a permanent arrangement for project road maintenance ; (c) the ptovisiou of additional incentives for laborers to perform, by relating payment to outputs e..g., by applying task work; and (6) encouraging the involvamant of the private sector, parti- cularly 8-11 contractors, i n labor-inte~ivework programs. Hitherto, they have been reluctant to mnage large labor forces, due t o the difficulties they perceive arising from Honduran labor laws, particularly regarding permanence of emplopmant, and minimum wages. 4.1.16 In general, the pilot project i n Bonduras can be considered as a successful operation since: (a) it has continued as a work concept in a properly established organization beyond the pilot stage (completed 1980); (b) the appropriateness of the labor-intensive work method has bean proven for the conditions pervailing in South-Western Bonduras; (c) this wotk method has been cost effective for most of the road sections selected; and (d) the program had a considerable socio-economic impact, although this has not been quantified by systematic field evaluations. 4.2.1 On the occasion of the f i r s t meeting of the SOL Project Steering Committee in April 1975, senior officials of the Kenya a n i s t r y of Works at- tended the meeting as observers. They expressed strong interest to apply the lessons of labor-intensive construction to the Rural Access Roads Program (RARP) of the Government of Kenya (GOK). The Steering Committee lent i t s f u l l support, t o the extent of suggesting that a portion of the planned experiments i n India and Indonesia could be curtailed t o accomm~datework i n Kenya. 4.2.2 The RARP was initiated when the GOK was confronted with growing unemployment i n the l a t e 196OPs/ear1y 1970's. The Ministry of Works was -- originally reluctant to use labor-intensive methods, c l a i d n g that they were too d i f f i c u l t t o implemcnt, -re too slow i n execution, required too much -Z superdsion, resulted i n poor quality roads, and ware not cost-effective. -ci The latter viw was supported by a consultant's report i n 1968. Bowever, the report also recognized that further studies w u l d be required t o determine the technology appropriate for road construction i n Kanya. Aa a result, the Internatioual Labor Office (ILO) set up a team, financed by the Norwegian Agency for Development ( N O W ) , t o collect productivity data for existing methods of road construction, between November 1974 and December 1975. 4.2.3 The work of the ILO/NORAD team was continued by an enlarged group of advisors, known as the Technology U n i t (TU), within the Roads Department of the Ministry of Works, sponsored by the Bank and the British Overseas Development Administration (OM) i n accordance with a Memorandm of Under- standing signed i n November 1975'by GOK and the Bank. The lQ started work i n January 1976, and according t o its term of reference provided assistance t o the management of the BARP i n devising ways t o u t i l i z e labor effectively. Ihe focw of the lQ assistance was on the implementation of improved con- struction techniques and the formulation of guidelines and procedures for - program planning, organization, and management of labor-intensive construc- tion. The work of three expatriate advisors was financed from a US$400,000 grant from the Project budget. The TU was supervised by the Bank with the assistance of ILO, which contributed technical assistance staff to RARP through a grant of the United Nations Development Programme (UNDP). 4.2.4 The TU remained i n existence u n t i l July 1978. In theory, i t was meant to have provided advisory services only, and its termination should not have affected ongoing operations. I n practice,-however, the activities of the TU also provided direct support i.a. i n supervision of construction, and i n the acquisition and maintenance of tools and equipment. The TU made significant c o n t r i b u t i o ~to the development of the program. Aoarever, it was placed outside the RARP administration a s a separate unit of the Special Projects Branch of the Roads Department (Figure 2); if it had been fully integrated under the BARP administration it could have had considerably nore impact; particularly i n solving problems as they arose. 4.2.5 During the period between October 1974 and June 1977 (the pilot phase), eighteen f i e l d units were t o be established under the RARP, each with about 270 workers. Bowever, only six had actually started work by June 1977. The delay was mainly due t o the shortage of headquarters s t a f f , a lengthy recruitment process t o r expatriate s t a f f , and delays i n the procurement of equipment and tools, a l l of which contributed t o a longer than forecast preparation t i m e t o s t a r t up the units. From mid-1977 onwards there was a steady build-up of units: by 1982, 44 had been established i n 24 d i s t r i c t s . with over 8,000 laborers employed. Ib of dd-1982, 5.885 km of rural access roads had been constructed and 2,223 km had been gravelled. In 1981, the average coat per km amounted to about US$ 6,400 for road construction in generally f l a t t o rolling terrain, with an additional US$1,600 for gravel- ling. While construction of x r a l access roads was gradually phrued out during 1984, progress continued with the gravelling of roada already built. By mid-1985, a t o t a l of about 7,640 km of roads had been constructed and about 5.250 im had been gravelled. z= 4.2.6 Ihe BbRP hu required c o ~ i d e r a b l emanagemant inputs from expatri- ates. Although the diract costs of the roads constructed was competitive, the overhaada and start-up costs, including technical assistance and train- ing, amounted to about 100%of direct costs in 1976. This acessive overhead load has diminished aa intensive efforts toward a major expansion i n the scope of work and Kcnyanization of management bore f r u i t ; in 1980/1981 the overhead costs had been reduced t o about 15 percent of diract costs. A t the beginning of the Program there * e r e no Kenyan engineers employed by the RAW, but gradually the number was increased, and i n 1984 only four expatriates worked under the Program. me large proportion of expatriate engineers employed i n i n i t i a l phases of the RARP should be seen i n the context of the Roads Department's general staffing situation during the 1970's. In 1977, of the ninety established positioas i n the Department, fifty-one were vacant and only twenty-one ware f i l l e d d t h Kenyans. Mded to t h i s had been the problem of the perception of a low status of the RARP by the local engineers. .Labor- intensive wthods were not regarded aa particularly attractive and wst of the positions involved residing i n the f i e l d , often i n out-of-the way places; t h i s was a further deterrent t o the scarce lpcal engineers, who preferred jobs i n the cities. 4.2.7 Considering the shortage of Kenyan engineers, training of super- visory staff became one of the important activities for the RARP. It was carried out effectively, and contributed t a the successful impleuentation of the program. Training started i n 1975 and since then, the Staff Training Department has managed, i n general, t o keep up with the requirements of the program for technical personndl.. Prom 1975 to 1982, about 40 unit super- visors (senior supervirors) and about 280 site supervisors (overseers) were trained, aa w e l l aa other staff for supervisory and operational activities (Figure 3). Training was, however, not provided for administrative per- sonnel. The importance of this wed was apparently underestimated by the Roads Department, resulting i n a serious impediment t o the progress of the Program i n 1977-78. The situation improved i n subsequent years due to continuous on-the-job training. 4.2.8 The supply of labor was generally adequate except f o r some seasonal shortages i n some areas during the coffee boom, particularly i n 1979. The MOTC resolved the problem i n 1980 by increasing the daily wage r a t e from KSh 7.9 (about USS 0.75) to KSh 13 (about US$ 1.20) for the program. Since 1980, the daily wage rate was only increased by two shillings to KSh 15 (about USS 1.40), in spite of a faster r a t e of inflation, without affecting the avail- a b i l i t y of workers willing t o join the program. A l l casual labor for the program w a recruited i n the immediate vicinity of the road constructed. When the supply of workers significantly exceeded the available jobs, a lottery system minimized the recruitment problems and eliminated the risks of favoritism and nepotism. 4.2.9 Zhe magnitude of the difficulties associated with tools was not adequately anticipated at the initiation of the RAW. For about four years, the lack of suitable tools hampered f i e l d work. Garden tools of inferior quality had to be used since the Supply Branch of the Ministry, in line with Covernmnnt policy, insisted on awarding tenders t o the lowust bidders, re- gardless of quality. The TU had to launch a f u l l exercise for tool design, testing and manufacturing. By 1977, the bulk of this work wan completed but another year was required t o modify the GOK procurement procedures to suit. The lcw availability of equipment also resulted i n poor f i e l d outputs, par- ticularly i n gravelling. According to a survey i n 1981, the availability of tractors and t r a i l e r s was only slightly above 50 percent. Equipment was procured too slowly and, once i n service, it was poorly maintained. 4.2.10 During the early stages of the program, various design standards were tried. Based on research carried out by the British Transport and Road - - Research Laboratory (TRRL) together with the TU and the kARP staff, a 4.50 m formation width (of which 4.00 m crown width were t o be gravelled for most of the selected road sections) was chosen. Since the sub-base was to k compacted by construction t r a f f i c alone, the camber was to be constructed a t about seven percent, reducing t o about five percent a f t e r compaction. Strong emphasis was placed on drainage to prevent erosion. Laterite gravel (murram) was used for surfacing, i n layers of 10-15 cm.. also compacred by construc- tion and users1 traffic. 4.2.11 The introduction of proper maintenance practises, particularly f o r these low-cost roads, is as c r i t i c a l to the long-term success of a labor- intensive program as construction itself. A number of alternative maincen- ance methods were considered, including the traditional method employed by the Maintenance Branch of the Roads Department. Finally. the Lengthman svm- tern was selected because it required only a small amount of equipment, waa simpler to monitor, and waa i n line with the principal concept of the pro- gram. Under this system, an average of 1.7 kn was allocated by contract t o a former construction laborer who resided close t o the road. Lengthmen are paid the equivalent of twelve days1 work per month; however, this is not re- lated t o the actual number of days actually worked; the principal requirmmat is that che road be maintained to a specified standard, except where major repairs are required.. Each road is inspected once a month. A survey conduc- ted i n 1982 found that almost 50 percent of the roads were i n good conditian. which is quite good cornpared with the record of the Maintenance Branch for roads of the lower classification. The annual total cost of routine mainto?- ance was i n average about USS200 per kilometer. As the need for additional new rural access roads wanes, the maintenance of the roads built under chis program w i l l increasingly become the principal activity of the RARP i n respect of unclassified roads. However, the organization and work methods ,f the BILBP are now k i n g wed by MOTC t o run a major program of rehabilitation of the classified road network (the "Mnor Roads Program"). 4.2.12 The labor-intensive work nmthod was i n i t i a l l y chosen for the PAD .. t o provide some relief t o the unemplopmant of the rural poor (para. 4.2.2.). To emure the long-term application of appropriate labor-intensive techno- -- logy, it was therefore necessary to demonatrate the cost effectiveness of this operation u 8 viable and swtdnable alternative to the conventio~l -=.-.?- equipsmnt-intensive c o ~ t r u c t i o nmethod. Since no comparative cost estimates were made a t the beginning of the program, 8 UK c o ~ u l t i n gfirm w commis- sioned to compare the coats of constructing the rural access roads and the tea roads i n bays. 3/ - The l a t t e r were built by contractors using equipment- intemive mathods. The financial comparison indicated that the labor-inten- sive operations, applied where feasible and practical, would reduce the t o t a l cost of the contracts investigated by an average of about 10 percent. Bow- ever, many questionable assumptions had to be made t o arrive a t t h i s figure and, as a result, the study was not considered t o have provided conclusive results. 4.2.13 Another attempt was made i n March 1985, when a joint Bank-fL0 mis- sion reviewed the cost and productivity data of the RAW. The mission found that the roads which provided the best examples for comparison wlth BdaP were those of the Gravelling, Bridge and Culverts Programme (GBC) which included the equipment-inrenaive reconstruction of classified D and E.roads. More . specificslly, the Kenya East Gravelling R o j e c t (KEGP) and the Central Kenya Gravelling Project (CKGP) '/ of this Rogram were anrlysed and the costs com- pared with the RARP. he-results are summarised in Table 8 which indicates the average direct cost and the average t o t a l cost per kilometer over the l a s t three f i s c a l years f o r the RAW, KEGP and CKGP. As indicated i n Table 8, the roads of the KECP and the CKGP are 64 percent and 92 percent respec- tively more expensive than those of the RARP i n terms of financial costs. 4.2.14 Although the r e l i a b i l i t y of these results is somewhat affected by differences between the RARP and the two projects of the CBC, the effect of these differences are significantSy reduced since they compensate each other, t o soma extent. Soma aspects justify the higher cost of the GBC project works, such as (a) the width of the carriageway, which for the GBC is between 4.5 and 5.5 m, compared with the RARP of 4.0 m; (b) the thickness of loose gravel (16 cm for the GBC, compared wlth 13 cm for the RARP), and (c) the level of co.mpaction (by r o l l e r s i n the GBC, and by t r a f f i c alone i n the -3/ Scott, Wilson, Kirkpatrick, and Partners. "Cost Comparisons of -Labour and Equipment-Eased Technologies ," 1977. -4/ The KEGP and the CKGP are sub-programs of the GBC. ,&UP). Conversely, the GBC reconstructs existing roads, while the RARP develops simple tracks into roads. Therefore, the amount of work necessary f o r bush clearance and rock/soil excavation is much larger for the RARP, and this would partly balance out the differences i n width and thickness of gravelling. - 4.2.15 It mat ba .-tad, however, that the validity of such a cost . comparison is uarully challenged by questioning the similarity i n geometric .-= designs, terrain and s o i l conditiow, dimate, the average haulage distance, incentives for laborers and supervisors, and the managerial/organizatio~l standards. Nevertheless, under the prevailing circuourtances, although the percentage cost differences may not be correct i n their absolute values, they support adequately one conclusion: the labor-intenstve mathod was not more expensive i n financial t e r m than equipment-based construction for the lower classes of road under the usual conditions i n Kenya. 4.2.16 The principal cost components vary significantly between the RARP and the GBC. A comparison of economic costs was therefore undertaken. As indicated i n Table 8 , the foreign exchange component of the GBC represents 48% of t o t a l cost, twice as much aa the samc component for the BARP. The mission estimated shadow price factors of 130 percent and 80 percent, respec- tively for the foreign exchange and local labor components. As indicated i n detail in the same Table, the economic costs of the GBC roads (built by equipment-intensive mathods) are about twice as high as those of the RARP. 4.2.17 Further socio-economic benefits vere identified by the monitoring and evaluation study financed by the COK and various aid agencies. The key findings of the study, based on a two-year survey of 828 households i n a selected impact area of the BARP included: - labor-intensive road construction succeeded i n employing primarily low-income residents, over half of whom earned wages for the f i r s t time; - the program created the means for marketing increased food crop and livestock production; - the rural access roads contributed to increased cash earnings from both farm and non-farm sources, and toward reducing the gap batmen upper and lower incoma groups. 4.2.18 The RARP also had an important development impact a t the local level due to the process through which roads were selected for construction. After some testing, the RARP adopted a policy of allowing District Develop- ment Councils (DDCs) to decide, uithin technical and financial parameters fixed by the Ministry of Transport and Communications, on the rural access road program for each district. This was the f i r s t concrete planning task s e t before the DDCs, i n whose deliberations the local politicians and t r i b a l chiefs were also invited to participate. This planning process later led to the preparation of full-scale District lkvelopraent Plans. The detailed discwsione of road construction program at the local level also paved the way for the recruitment of labor and the identification of local sources o f . materiala and supplies t o carry out the construotion work. 4.2.19 In conclueion, the pilot project and the subsequent implamantation of a full-scale program on rural access road c o ~ t r u c t i o ni n Kenya demon- -?- strated the importance of: C_ - ~ (a) the COK's commitnant and positive a t t i t u d e towar& labor- intensive construction technology, and the continuous support by various aid agencies (although the origiml'program for constructing 14,000 bn of such roads over five years 'was too ambitious, the BBRe actual output of about 7,000 km of roads from 1975 to 1984 still represents an outstanding achievement for t h i s type of work); (b) setting up appropriate institutionslarrangemants f o r such a program: the Enads Department created a Special Branch'as a central source of support, while the impleumatation was decan- . tralized including, a t the planning stage, the involvement of the D D 6 ; (c) the determination of t h e appropriate work method for low-cos c roads providing all-weather access to r u r a l communities and. i n particular, designing engineering and management procedurar especially adapted to a large-scale program in a country vhare labor-intensive construction has not been used traditionally; (d) labor-intensive construction and maintenance methods for che creation of employment opportunities; i n FY1982, the labor force of the RARP consisted of about 9,600 workers for con- struction and about 3,700 for road maintenance, of primari:~ low-incormr residents; e staff training, since labor-intensive operations require supervisory and management staff with special a b i l i t i e s co manage large numbers of unskilled and often i l l i t e r a t e workers; and ( f ) training of administrative s t a f f , which was underestimatad i n the early stages of the RARP implementation: the need for specially trained store-keepers, for instance, was not adequately considered initially. 4.2.20 To sum up, Kenya's RAW can be considered as an outstanding achievement, i n terms of the mileage of roads contructed; the institutional integration of the RAW; the special arrangements for training technical s t a f f ; the cost effectiveness of the operation a f t e r an i n i t i a l learning period; aqd its socio-econocnic impact. The knowledge and experience gained i n a n y a could be applied to similar projects elsewhere and was, therefore, studied by representatives of other developing countries as an evidence that labor-intensive construction methods can be viable and competitive. The impressive results of the BARP encouraged the GOK t o widen the scope of . applying labor-intensive emrk methods by wing them for the "Minor Roads Pro- grame" on which MOTC has recently embarked. Thus, although the RBBP encoun- tered significant difficulties at the initial stage, it was later established aa a viable organization within the Boa& Dapartmant of Hl'OC. The RARP de- -- ' -< monstrated that a large scale program of lab~r-fntensivew r k can be imple- \ mented affectively. Bank-sponsored Programs i n other Countries 4.3.1. - The Bank has also sponsored labor-intensive c i v i l r o r k a programs i n countries other than those indicated i n the previous chapters of this report. The SOL Project, however, did not contribute financially to those programs but supported them by advice and technical assistance which was rendered by members of the study team. The special experience gained with the pilot projects in Chad, Benin, Lesotho and the Dominiclan Republic are highlighted below. 4.3.2 I n s , the Government had undertaken a large irrigation program i n the southern part of the country i n the early 1970's. The f i r s t phase consisted of the Sategui-Deresia Irrigation Project, which aimed to level and irrigate about 4,000 ha for rice cultivation. Financing was provided by the Government of Chad (GOC), the International Development Association (IDA) and the African Development Bank (AFDB). The GOC decided to carry out the c i v i l w r k s by force account using equipment-intensive methods under the management of an Italian consulting firm. While awaiting delivery of the newly-pur- chased equipment, the project authority started construction work i n 1976 with a few pieces of equipment borrowed from other governmental agencies and with a labor force of about 500. During this period, the labor productivity was well below the expected level. This was largely due to the use of agri- cultural tools, unsuitable for construction work; poor organisation of the work; and a general la& of site supervision. Hovever, the positive expe- rience of having been able t o s t a r t up construction work with a m i n i m equipment f l e e t and making intensive we of unskilled labor prompted the GOC t o s e t up a labor-intensive pilot project to investigate the viability of t h i s technology. A consulting firm from the United Kingdom (Ug) was selected to provide ona expert for assisting i n the implementation of the pilot pro- ject. The project was started up i n 1977 on a modest scale, to work out the details of organization and logistics, train supervisory s t a f f , and determine a t what productivity rates labor-intensive work in Chad would be cost effec- tive. The GOC was satisified with the results obtained i n the pilot project. and decided to continue and expand the project i n 1978, and to prepare a long-term labor-intensive construction program. However, after the 1978 working season this project was suspended when c i v i l unrest turned into open warfare. 4.3.3 The labor-intensive construction experience outlined above was a relatively small s i z e experiment which, however, yielded several useful lessons, most of which confirmed the experience i n other countries: (a) The introduction of a large-scale labor-intensive construction operation i n a country without prior experience of such tech- nology requires the assistance of qualified mnagers, usually expatriates. The i n i t i a l overhead cost is usually high (the g- cost of the UK consultant amounted t o 65 percent of t o t a l costs in 1977 and 29 percent i n 1978). (b) The selective use of labor-intensive construction i n irriga- tion works can be cost effective. Labor and equipment can form a least-cost combination, even if labor is available only seasonally and a t a wage r a t e of about USS1.60 per man-day (1978 prices -'/). (c) Local laborers were involved i n the construction of those works (e.g. small irrigation canals and enbankments) which, a f t e r completion, they (as f e n m r s ) would use and would have t o mrfntain with a technology appropriate to their mans. Furthermore, the opportuaity of securing a cash income during a period when those farmers were normally not employed (up to 50 percent of the year), was very uaeful. 4.3.4 In the early 1970's the r u r a l roads network was poorly developed i n the People's Republic of Benin. With IDA assistance, a feeder road construc- tion program was prepared i n 1975. including 700 km t o be constructed. 600 km of secondary roads fc be regravelled, and 3;300 km to be maintained. -The Government proposed to carry out the work by equipment-intensive "brigades" which were to construct or improve 125 km per year and unit. The use of labor-intensive or other appropriate construction methods was f i r s t consid- ered when IDA appraised the program as part of the F i r s t Feeder Roads Project i n July 1976. The appraisal team recommanded, and the Government of Benin (GOB) agreed to, the creation of oae equipment brigade, one intermediate brigade (which sought t o define an optimum labor and equipment mix) and two labor-intensive brigades. The work was carried out by a newly created Feeder Roads Division (Division des Routes de Desserte Rurales, DRDR) of the Ministry of Public Works, Consrructioa and Housing. A UK consulting firm was retained t o provide management and technical assistance t o the DRDR. The experiance gained with this Project (May 1978 to January 1981) and with the Second Feeder Roads Project (appraised i n Nay 1980) provided an opportunity to compare the performances and cost of labor-ineensive with equipment-based construction methods under similar circumstances. -5/ Francs CFA (FCFA) 230 (about USS0.94). pre-July 1977 FCFA 250 (about USS1.02). July 1977 to January 19i8 FCFA 365 (about USS1.62), a f t e r January 1978 4.3.5 The composition of the brigades is illustratad i n Tables 9 and 10. These tables clearly indicate the differences i n equipment value (about 4 t o 1) and i n number of man-onths of unskilled labor employed (about 1 to 4.75) for equipment versus labor brigades. The daily wages ware FCFA 355 i n 1978, FCFA 385 i n 1979, and FCFA 450 from 1981 onwards (equivalent to about USS1.70 a t the 1981 exchange rate). The major factors affecting the produc- t i v i t y of the equipment-intensive and intermadiate brigades were the avail- a b i l i t y o f skilled operators,. the maintenance of the equipment, and the supply of fuel and spare parts, while the labor-intensive output depended very mch on the level of labor management 'and the leadership exercised by the expatriates and local field supervisors. 4.3.6 The average annual construction targets under the i n i t i a l project were set a t 50 km for the labor-intensive, 75 km for the intermediate, and 125 km for the equipment-intensive brigade. The appraisal mission had esti- mated that a t these annual production rates, per-kilometer costs of road improvements could be approximately the same. After about a year of opera- tion, output rates by the labor-based brigades were on target. The inter- mediate brigade attained about 75 percent of the estimated output, largely due t o the difficulties inherent i n organizing mechanized works i n conjunc- tion with labor-intensive ones. The heavy brigade achieved only about half its target output. The shortfall of the l a t t e r wns mostly cawed by problems with the supply of fuel and spares; the highest annual output it achieved was 87 km i n 1980, when the equipment was still f a i r l y new and many of the - initial start-up problems had been resolved. In 1982 the situation deterio- rated t o the point where production came t o a s t a n d s t i l l because most of the heavy equipment ueeded repair. Since then, the mechanised brigade has continued to suffer from problems of lw equipment availability, and it may be phased out during stage 111 of the feeder road programs (1984-1987). 4.3.7 men the second phase of this program was appraised by IDA in the spring of 1980, the intermediate brigade was still performing poorly. Conse- quently, it was planned to phase it out and use the equipment to form a main- tenance unit. However, the coordination of equipment and labor operations improved subsequently; output. significantly increased by mid-1980; and dis- bandment of the unit was deferred. 4.3.8 The labor-intensive brigades gave the most consistent performance, w i t l ~an output of about 40 km per year and unit after 1979. As a result, four more brigades were included i n phase I1 of the feeder road program (two financed by ILlA and two by the United Nations Capital Development Fund). The brigades operated mostly on roads where there were comparatively short hauls for earthworks and gravelling, so as to nake economic use of the combination of agricultural tractors and tipper trailers. Productivity incentives were established, with daily tasks set for each labor gang based on average individual outputs, and the gangs setting their own working pace. - 4.3.9 Bued o a the data gathered from 1979 to 1982, the approximate average coats (in US$) 6/ per kilomater of road a r e summarized as follows: Plus: Method Mract Indirect Subtotal Techaical Total .. .......A8sintmce . . Equipment-basad 7,625 1,690 9,314 1,338 10,652 Labor-intensive 7,783 1,060 8,843 1,523 10,366 4.3.10 The direct costs comprise the expenses directly attributable t o a particular s i t e , including equipment depreciation, but excluding other costs, such as interest on capital, which would affect the mechanized option more markedly. Proportional headquarters expenses and training costs are covered i n the "indirect" category. The technical assistance comprises the cost of tha axpatriates who'would eventually bs phased out when qualified Bednois personnel cm take over. By mid-1983, the DRDR personnel amounted t o nearly 2,000 including 1,500 laborers, but substantial technical assistance was still required due to the general shortage of qualified local staff i n the c i v i l service. 4.3.11 Unfortunately, the data used to calculate the average costs were not completely compatible because the mathod of equipment depreciation and the allocation of costs was changed during the recording period. Further- more, cost accounting did not directly r e l a t e t o physical progress. The above comparison which indicates an about five-percent higher (direct plus indirect) cost for the equipment-based brigade, should also be seen i n the context of the fluctuations which occurred i n the relative costs during the reporting period. This fluctuation ranged from about 13 percent lower t o about 42 percent higher costs of the equipment-intensive brigade compared to the average of the labor-based units, excluding those assisted by UNCDF, which started only a t the end of this period. -6/ Exchange rates used in this report: Calendar Year US51 = FCFA Calendar Year USSl = FCFA 1977 245 1980 215 1978 225 1981 265 1979 210 1982 340 4.3.12 Tha Banin experience is particularly valuable for those officials in developing countries and i n aid agencies who wish t o test the different work mathods under specified conditions. An important lesson gained i n Benin was that the mathodology of a cost comparison should be agreed upon i n detail prior to the s t a r t of work, and the basic assumptions should remain the same throughout the reporting period. Furthermore, the mthodology used f o r recording and modtoring the costs should ba adequate, yet simple to ensure proper implemntation and consistency. Finally, the mthodologg should focus on unit costs of key construction activities, such as excavation, hauling and collpaction, rather than on a global per kilometer rate, t o provide a better basis for cost comparison. 4.3.13 Nevertheless, i n the Benin experience, although the basic informa- tion used t o derive the average costs per kilometer of construction shows inconsistencies along the period of execution of the work, some of these in- consistencies are mutually compensating. The average figures indicate that the labor-intensive methods are competitive with mechanized construction i n financial terms. It can be further concluded that the labor-intensive methods (a) are also economically competitive, (b) are relatively insensitive t o the supply and repair problems of the mechanized brigades, and (c) have the additional advantage of returning a considerable amount of construction Coats i n wages t o the local economies the roads are going t o serve. 4.3.14 The conditions i n the Kingdom of Lesotho provided another valuable t e s t wse for the use of labor-intensive work methods. The~economyof iasotho depends t o a large extent on the availability of employment oppor- tunities i n the Republic of South Africa (RSA). In the early 19701n, about sixty percent of the able-bodied male population (about 175,000 man) were working as migrant laborers i n the RSA, compared with 20,000 total wage earners i n Lesotho. The average earnings of the migrants was estimated t o exceed the Gross Domestic Roduct (GDP) of Lesotho. In 1974, disturbances i n the mines of the RSA resulted i n the repatriation of about 15,000 migrant laborers. Similar situations could arise in the future; this factor, together with an annual population growth of about 2.3 percent, point a t the urgent need to increase domestic employment opportunities. 4.3.15 The Government of Lesotho (GOL) considered varioua emergency c i v i l works projects i n 1974 t o occupy the repatriated workers who were unable to return to their farms. However, COL soon discovered that there were no suit- able institutional arrangemcnts nor expertise i n the country t o administer and supervise such a a major undertaking. The Government therefore arranged, i n cooperation with the UNDP and the Bank, a study to identify broad strate- gies to increase long-term employment opportunities and to prepare an action plan. The study report of April 1975 recommended the establishment of an experiemental labor-intensive construction unit and a national employment service. The GOL implemented both. A three-year pilot operation for the Labor-intensive Construction Unit (LCU) was arranged under the IDA Second Highway Project, approved i n 1976. The LC3 was made into a special unit within the Ministry of Works reporting directly to a Coordinating Committee of the GOL. Because of the general shortage of qualified indigenous s t a f f , the three senior positions in the LCU were to be f i l l e d Sy a team of expat- riates. In 1977, operations started with road works financed by grants from ODA and the Swedish International Development Agency (SIDA). The COL finan- ced additional road uorks, conservation works, construction of f i s h ponds, regravelling of air strips, land terracing, etc. In the initial stage, so- 200 t o 300 laborers ware employed. 'Ikese initial LCQ operatiom were c o ~ i - -x- dared successful, and the technical wsistance waa continued under the IDA Third aighway R o j e c t , approved in 1979. 4.3.16 Howaver, when the IkU operations were extended, and the number of laborers amploped increased to over 2,000, problems started t o arise. The i n i t i a l mini- wage of U1.60 (about USS1.86) per day was increwed t o M2.40 (about USS2.84) i n May 1978 and to U3.40 (about USS4.10) i n April 1980, '/ without a corresponding increase i n productivity. Ihis affected the cort- effectiveness of the operations. Lack and frequent turnover of supervisory staff also resulted i n lower outputs and quality on some sites. 4.3.17 The problem of supervision were complicated by the f a c t that, ta meet -the GOL's ultimate goal of creating ethployment for many thousands of repatriated workers, the LCU was used as a training ground f o r soma key personnel of other Ministries and Departments. Supervisory staff from the Comervation Division of the Ministry of Agriculture and the Boa& Division of the Ministry of Rural Development have received training t o increase the productivity of their Food-for-Work teams. Zhe Water Branch of the Ministry of Wnes, Water, and National Rasources has adopted LCU nmthods and has experienced a significant increase i n direct labor productivity. Kowever, the training of local supervisory staff for the LCU and other government agencies has not kept pace with requirements. 4.3.18 The composition of the hired labor force varied from agency t o agency, even though the recruitment of laborers was carried out by all agencies with the cooperation of the local authorities, and an adequate sup- ply of labor was always available. Male laborers comprised the labor force i n the LCU, while the Food-for-rk program of the Ministry of Rural Develop- pent hired a majority of women. The explanation for this difference might be found i n the traditional, patriarchal society i n Lesotho; men play the domi- nant role i n the villages which are the basic political aad administrative unit. Women usually spend their time performtng either domestic duties o r -71 - Lesotho Maloti (H) 1.00 RSA Rand (R) 1.00 USS1.OO = M0.86 i n February 1976 = N0.83 in January 1980 = M1.20 i n April 1984 agricultural wrk, i.e., producing food. Therefore, when food was used to pay for wrk, wmcn.were aomi~atedby the village chiefs. Uen were available for w r k remunerated i n cash. 4.3.19 Although the GOL's original goal wa8 the employment of a very large labor force, the need to aim f o r cost-effective operations was dictated by the limitation of available resources and competing demands for those resources. 'Iba significant increase of the minimum wages between 1975 and -- 1980 required a review of the viability of labor-intensive work methods com- pared with mchaaisation, using equipment offered at highly competitive rental rates by specialized enterprises of the BSA. To this end, the costs of similar improvemnta for tw roads, one executed by labor-intensive and the other by equipment-intensive methods were analysed. Similarities between the two roads existed also i n the proximity t o the quarries or gravel p i t s , and i n the duration and timing of construction i.q., about four months i n 1982. Although the analysis was not finalized for lack of soma key data, it clearly indicated certain significant aspects i n the comparison of the performance between the equipment- and labor-intensive brigades; the,former compared to the latter- - completed 27 km versus 16 km; - produced a good quality of road, suitable for high t r a f f i c density versus a quality suitable for only a low volume of traffic; - due to the higher quality of road, cost si nificantly m r e per km than the other, although the cost per m5 of gravel brought . on to the road was about the same i n the two cases; - is very sensitive to distance between the 'site and the supplv depot versus largely insensitive t o remoteness; and - requires only conventional work arrangements versus the need for proper planning and management of the logistics for a larre body of workers. 4.3.20 Guidance of a somewhat different sort was provided by the Bank t > the Dominican Republic. The economic mission visiting the country in November 1976 included tw staff members of the SOL Rojecr, who assessed I?. feasibility of instituting a program of labor-intensive civil constructioa. A t the request of the Government (GODR) a seminar was held i n Santo Domine9 i n October 1978 by the two Project staff members, presenting the Bank's experience with labor-intensive work methods. GODR's interest was aroused because of three principal reasons: a ) the potential for savings of foreign exchange; b) .nopportunity to enhance ?ocial cohesion in the rural areas due t o the involvement of the comrmnity i n the construction of its & road, and the attendant possibility that the road would l a t e r .be maintained by the same commaity; and c) the potential creation of jobs. 4.3.21 The Bank miasion appraising the Second Road lhinteaance and &.con- a- struction Project in June 1979, established that labor costs were c r i t i c a l to the introductioa of labor-intensive construction techniques id the Dominican public. The economic analysis showed that, at a shadow rate of BD$2.008/ per day, which corresponded t o the shadow wage r a t e f o r rural umkllled lzbor i n the country, the economic cost was similar for labor and for conventional equipment-intensive techni ues. At the minimum legal daily wage at that time for r u r a l labor of 80$3.50 1 , the financial cost of a given road section was 1 up t o 20 percent higher apFlying labor rather than equipment-intensive techniques. The CODR waa prepared to incur this additional cost to test the technical feasibility of labor-based techniques f o r the construction of roads. kr the GODR1s view, the additional cost was j w t i f i e d by the benefits from employment and income generation i n selected r u r a l areas. 4.3.22 The lack of adequate and timely budgetaq support, largely cawed by the emergency situation following a devasting hurricane i n l a t e 1979, cawed excessive delays i n the payment of wages, which reduced productivity, increasing unit costs beyond an acceptable level. To eliminate these cash flow problem, the Bank increased the disbursement percentage from 63 t o 100 percent under Loan 1784-DO i n May 1984. With payments to workers secured, the labor/equipment mix used under the pilot project has been reevaluated t o apply only cost-effective work methods for the construcrion and improvement of rural roads under the Third Road Reconstruction Project assisted by the Bank. Nevertheless, the f i n a l report of the consultants, issued i n July 1985, concluded that the cost of labor-intensive construction exceeded that of equipment-intensive methods, and questioned the continuation of the program. 4.3.23 An important lesson emerged from the experience i n the Domini'dan Republic: since laborers must be paid i n a timely manner t o maintain an ade- quate level of productivity, the success of applying labor-baaed work methods depends very much on the regularity of cash flows. The opportune provision -8/ - Peso (RD$) 1.00 USS1.00 -91 The dnlmum wage had been raised from RDS1.50 to RDS3.50 in several steps from the time of the v i s i t of the SOL Project team i n 1976 to the time of appraisal. of fuel and spare parts (also dependent on. cash flows, La.) also affect equipmant-intensive operations. 'he principal difference between the two methods is i n the time elapsed before the impact of delays i n cash avail- ability. While the non-payment of laborers affects productivity and causes adverse political effects locally almost imadiately, the shortages of fuel etc., reduces the output of equipment-intensive operations, but the public is apparently more willing to accept this a# unavoidable. Labor-intensive operation8 have obviously a higher political sensitivity than the equipment- ----- b e d mthod, and soma governments are perceived aa being reluctant t o commit tharmelves t o labor-intensive operation^ i f a direct support by an aid agency haa not been secured. 4.3.24 In addition t o the experience i n India, Indonesia, Honduras, Kenya, Chad. Benin. Lesotho and the Dominican Re~ublicmentioned above. the Bank has supported the revival or introduction of iabor-intensive work Athods i n many other developing countries such as Malawi, Botswana. Burma Philippines, -1 Colombia, and Ecuador, as an immediate follow-up of the SOL Project. Malawi and Botswana benefitted from the emerience gained i n Kenva. The traditional use of labor has been made more efficient in-BU- for the work of the Con- struction Corporation and i n the Philippines for the construction of the barangay (rural access) roads. The labor-inteyive (Pico y Pala) road program was supported i n Colombia, while i n Ecuador the traditional minga system was reactivated for an irrigation project. Conditioas and require- ments vary from country t o country but not the principal goal, that is t o make economic use of. the resources available i n a country including the most important one, the human resource with focus on the untapped sector of un- ehloyed or underemployed, unskilled labor. 4.3.25 The work of disseminating appropriate labor-intensive construction and maintenance methods continues, and the list of countries i n which the Bank is supporting related interventions is continually expanding. Likewise, efforts are being made to extend the sphere of applications t o a wider range of rural and urban projects. The coordination of efforts with the ILO and the UNDP is assisting considerably i n this respect. 4.0 hograms without Bank Support 4.4.1 lbe pressing need to create jobs for unemployed laborers has forced many governments of developing countries to try the application of labor- intensive work methods i n public works program even without the support of external agencies. There have been frequent failures, mainly due to the lack of planning and availability of qualified supervisory personnel. An excep- tion, generally consi'dered a success story, is the rural road construction program i n Mexico, which was reviewed and reported upon as part of the SOL Project. 4.4.2 During 1971-83 about 80,000 km of rural access roads were built in Maxico under t h i s program, mostly by labor-intensive technology. Initially, laborers were employed in preference to equipment only as a means of creating work opportunities for economically depressed agricultural areas. Later experience showed that, given r e a l i s t i c factor costs and a suitable orgadza- tion, labor could compete favorably with equipment f o r rural road construc- tion. Other requirements were that specifications should be appropriate to labor-intensive construction, and t h a t work should be carried out i n terrain which lent itself to pick and shovel technology. The development of labor- intensive work methods i n Mexico is perticularly noteworthy because: - It brought new awareness- to the rural commrdties of their cap- --=- -% a b i l i t y and strength t o build works f o r their own m l f a r e and t o employ the collective efforts of commnity members i n the development of infraatructure. - It offers an excellent example of the development of para- professional human resources needed f o r labor,+ntensive work, and of how d v i l engineers can be encouraged to espouse this technology as a worthwhile professional pursuit. - It generated a well-established Government institution. 4.4.3 In 1972 the Directorate-General of Rural Roads Construction ( W R ) was established, ranking as a full-fledged department of the Ministry concerned with transportation. The f u l l support given by the Federal Govern- ment and particularly by successive Ministers was a key element i n the de- velopman~of labor-intensive technology i n Mexico and of an institution which has gained permanence within government. DGCB has been able (a) to operate at grass-roots level and involve the c o m n i t i e s i n self-help construction work; (b) to develop a large body of professional and, particularly, of paraprofessioaal s t a f f , and an fngrained, intense esprit de corps and pride of service; (c) t o provide a good "visibility" f o r staff promotions within the Ministry; and (d) to apply design characteristics and specifications 'which are suited to the t r a f f i c levels and the s i t e conditions prevailing t? Mexico. Another important factor contributing to DGCR strength has been i t * decentralized approach t o management. DGCR has delegated authority to "Residencias Generales" (General Residencies) i n each of the Kexican staces which manage the planning, design and execution of rural road construction within their jurisdictions, and coordinate the work of special programs f o r the development of rural roads (Figure 4). 4.4.4 Although communities have traditionally organized themselves f o r carrying out emergency and minor iafrastructure works f o r the common good. there was no tradition of coomunal n r k i n road construction or amintensnco Organizing the communities into building their own roads was also necesrar- to achieve cost efficiency and extend the reach of h d g e t funds i n two wavs. by obtaining (a) a l l local materials and the right-of-way from the commun- i t i e s , and (b) higher productivity through communal caskwork and the re?u-::- ing peer pressure amongst individuals, t o achieve community objectives. " In this respect, the communities became DGCR "contractors" working on aT:?y - 101 The Nexican Constitution prescribes.againsc piece work by individuals and the differentiation of payments i n accordance with productivity. of lump sum contract. .In addition, communities would later help in maintaining their own access roads, d t h the Ministry carrying out major repairs or supplying equipment when required. 4.4.5 M;CB therefore developed a strategy whereby the communities are i n i t i a l l y approached by specially trained "promoters". Zhese are para- -- ptofessionals who mst have tact, sensitivity, and ability to communicate effectively with, and organize. the rural commupities. Aa a f i r s t step, the -.= promoters survey the communities in a region t o determfne priorities for action by DGCR. Once a specific road has been selected for development and the villagers have agreed t o participate i n its C O M ~ N C ~ ~ Othe promoters ~ , assist the community leaders, whose authority they respect and through whom they act, to establish a Works Committee formed by members elected amongst the villagers. Ihe Works Committee then enters into an agreemnt for the construction of the road with DGCB. Typically, DGCR (a) provides road design, technical and administrative supervision, (b) supplies equipment, tools and manufactured materials, (c) provides the funding for wages cal- culated on the basis of DGCR standard productivities, and (d) introduces trained, experienced candidates for foremen. The l a t t e r and a l l the labor are employed by the Works Committee, which also supplies local materials and arranges all rights-of-way and easements. DGCR helps i n a l l techni,cal aspects including setting up a system for timekeeping and payroll prepa- ration. However, labor disputes must be resolved by the community, as the direct employer of a l l labor. 4.4.6 DGCR has dedicated considerable efforts t o the development of per- sonnel particularly suited to the construction of rural roads with minimum resources, using appropriate technology. Zhe engineers, for example. had t o be motivated to accept a less glamorous technology, and work i n the poorer and more isolated areas of the country. They were encouraged to espouse labor-intensive rural access road construction by reasoning that, for build- ing a highway, considerable technical and financial resources cust be made avdlable, and the individual engineer's efforts are usually engulfed i n the activities of a large team, with l i t t l e perception of the individual's con- tributions to the overall results. In contrast, for building a rural road with scant technical and financial resources, the personal ingenuity, cre- ativity and perseverance of the individual engineer are put t o strenuous tests, and the results of his efforts are readily visible in the inproverent of access to hitherto neglected villages. These results represent immediate professional rewarde. As mentioned previously, this approach through prose- lytization was supported by the establishment of a career pattern for DGCR s t a f f , and by the permanence of DGCR i t s e l f within the structure of the Ministry. Nevertheless, there was a high turnover of engineering staff i n the i n i t i a l years (about 50 percent per annum), which l a t e r settled to levels comparable with those of other offices of the Ministry. 4.4.7 Over the period 1971-82 the rate of use of labor fluctuated largely i n response to political and econodc circumstances. After the i n i t i a l growth period, it peaked i n 1973, a t about 60 percent of overall costs. hrring most of this period, the wage r a t e was possibly below the interna- tional threshold for labor-intensive construction cost efficiency, estimated f o r 1975 a t about US$2.00. By l a t e 1973, even the 1975 threshold had been exceeded, and by l a t e 1974 the wage r a t e was almost 1.5 timas the 1975 threshold level for construction workers and 20 percent over f o r farm labor. This brought about the introduction of a polley to raduca labor intensity i n rural road comtruction. 'his policy was reversed i n 1975, vhen the use of s- labor w8a again encouraged. A new peak of labor content (about 63 percent of overall costs) w8a reached i n 1977. Ovar the 1971-79 period, r u r a l road construction in Mexico was therefore predominantly labor-intensive, labor inputs having averaged about 50 percent of t o t a l costs over the whole period. From 1979 t o 1982 the Mexican economy was buoyant and rural real incomes soared. Labor participation i n the rural roads program dropped steadily, reaching about 15 percent i n 1982. The financial c r i s i s which developed i n 1982 once again affected the r u r a l economy and reduced the r e a l wage levels of agricultural labor. This factor, combined with the poor harvest obtained i n 1982, promted DGCR t o reembark on labor-intensive construction. The new labor-intensive programs are being implemented once again, where feasible, i n collaboration with the local commrnities i n the poorer areas of the country (e.g., i n the State of Chiapas, where the Bank is now supporting a rural roads construction project). 4.4.8 In considering the replicability of the Mexican experience, two points should be stressed: f i r s t l y , that the application of labor-intensive construction is very sensitive t o the cost of wages, as witnessed by changes of policy of DGCR and by the d i f f i c u l t i e s experienced by contractors to apply labor-intensive technology; and secondly, that the Mexican example has developed i n an environment where c i v i l engineers and trained technicians are abundant and the general literacy level is relatively high. This has assured a good supply of suitable candidates for special training as professionals and para-professioaals from which t o develop DGCR's organization. 5. FINANCIAL EVALUATION OF TEE CONCEPT. AND ECONOMIC ASPECTS 5.1 This rasearch project wae concerned with the evaluation of a work meth~d,as opposed t o a specffic economic asset where the economic rate of return could have been evaluated by standard techniques. Efforts were therefore made to establish comparisons of cost between labor- and equipment- intensive c o ~ t r u c t i o nmathods, as a means of determining the appropriate .technology. This task has proven d i f f i c u l t , as incomplete information fro= ' force account operations and price distortions cloud the true comparison. Nevertheless, the f i e l d work under SOL produced some indicative results as detailed under Chapter 4 and a t least one conclusive cost comparison study for the Honduras program. This chapter discusses some of the problems encountered, and offers a few pointers for future cost comparisons. 5.2 Tha initial enthusiasm shown by many governments for the intro- duction of labor-intensive technology as a means for easing unemployment has often k e n dampened by doubts regarding the cost effectiveness of the labor- intensive operation. Governments are usually confronted with demands from l i n e ministries which exceed the financial resources available. The question has therefore to be raised, whether labor-intensive work methods are more expensive than conventional equipment-based operations for the c o m t ~ c t i o n of civil MI^ and, consequently, whether they require additional funding. -- A prudent financial evaluation of the concept is therefore required t o ate- -?-. gorire labor-intensive work mathods either for the use i n social program (i.e., higher costs are counted agaimt social benefits such ao employment and improved incoum distribution), or as a feasible alternative to equipment based operatiom. 5.3 Accurate and unbiased factor inputs i n meaningful cost acounting and recording systems are necessary t o properly analyse the costs of the two construction methods. When accuracy and relevance of recording varied signi- ficantly between projects reviewed i n this report, the reconciliation of recorded data gave rise to errors which could affect the conclusions of cost comparisons. 5.4 The achievement of a true cost comparison between the tw work methods also requires that the operations to ba compared be carried out under normal conditions. Such conditions would include, for example, that: (a) laborers be paid promptly, and fuel and spare parts be available for the equipment; -(b) operatious have passed the introductory (or pilot) period and produce routinely a t a generally accepted level of quality and efficiency ; and (c) prices reflect relative scarcity of resources e.g., the currency is not overvalued and wages are a t market rates. 5.5 In reality, these "normal" conditions may not a l l be present i n every case under scmtiny. Nevertheless, some of the requirements to nor- malize conditions can be met prithout undue effort, for example, the timely provision of funds, the availability of fuel and spare parts, and that operations have passed the "learning" phase. The elimination of price distortions, however, requires difficult policy decisions, which often prove t o be insurmountable obstacles. They uust therefore be dealt with through the evaluation of economic costs. 5.6 The specifications for construction used i n industrial countries and usually adopted i n the developing world assume an equipment-intensive execution. Therefore the comparison between the work methods may entail the comparison of tw different results achieved under different specifications. These differences nay be related to the duration of the construction period, the economic l i f e of the completed works, the maintenance costs and users' benefits, which in many situations may be less favorable for the labor- intensive works. Common denominators can be found for the economic signi- ficance of alternative technologies, but this elaborate evaluation concept .' has rarely been considered because it is time consuming and requires making various potentially arbitrary a8sumptions. 5.7 The attempts aade i n several countries to compare the coats of the alternative construction lnthods were uaually limited to comparisons of cash flowm i.e., the amounts of liquid funds spent fbr mch method. W l e the direct cost of labor (but often not the overheads) can easily be identified from payroll#, the cost of equipment is uaually dlstorted by unrealistic assumptions, or omission, of cost factors such as the true cost of owning and operating equipment. Equipment donated by external agencies is wually not considered by the governments concerned as owner's costs i n a general comparison of work methods. The cost of technical assistance by expatriates may either be omitted altogether, or incorrectly analyzed, if the learning period (during which technical assistance is heavy) is not considered sepa- rately from routine execution periods. A certain amount of duplication may also be involved here, in the cost of local staff who work as understudies of the expatriates. 5.8 The distortion of factor inputs i n a cost comparison can be even greater when force account operations are compared wlth contract work. The fact that the contractor has to baar a price risk is ususally not considered when comparing two projects with different types of executors. The force account unit would receive additional budget allocations in the case of a cost overrun; the contractor has to include the likely additional costs in his bid price, or risk a loss. Even when the actual cost of the wrk performed by a force account unit is compared with a final contract amount based on unit prices, the result may not be representative since it i s unlikely that performances were average in both cases. 5.9 The technique of economic evaluation, where price distortions are involved, usually entail the w e of shadow prices. This involves primarily weighting labor and/or foreign exchange coats against actual market con- ditions. In practice, however, the comparison of financial costs is far morc convincing to governments in the Third World than figures based on economic readjustmant. The economic cost comparison may, nevertheless, become a decisive factor when the results of the financial cost comparison do not clearly favor one of the alternatives. 5.10 Perhaps the closest approximation to a true and compr,ehensive comparison of financial costs was carried out in Honduras, where several rural access roads programs were executed between 1982 and 1984, using various kinds of procurement. The conditions in Ron.duras were favorable for t h i s type of analysis. Datailed records were available for the force account operations, and they were compared with the final costs of contracted-out work. This i s the only case known where the thresholds for .labor- and equipment-intensive work methods have been identified with some degree of reliability, and where conclusive evidence of the relative costs of executing the work by the alternative methods has been obtained. 5.11 In summary, the experience gained i n comparing the costs of labor- wtth equipment-intensive construction showed that there is no clear, world- wide sweeping advantage of either. Each case uust be considered on its own merits. There are many instances when labor-intetwive cowtruction methods dl1ba financially more attractive than mechanized co~truction,parti- _-.--- cularly i n countries where the prevailing agricultural daily wages are below L> empirically identified thresholds. In cases where there is a financial advantage for equipment-intensive operations, an snalysis of economic costs w i l l often show the convenience of employing labor-intensive methods. However, governments are inclined to choose construction methods on the basis of financial significance alone. 5.12 Government officials of a developing country, where no experience exists, or where unsatisfactory results were previously achieved, with large- scale employment of unskilled labor i n construction, might be confronted with the question, whether or not to consider or reconsider the we of labor- intensive construction methods i n executing c i v i l engineering works. A proposed system for the screening of prevailing conditions i n a country is outlined i n Figure 5 and for application t o programs/projects i n Figure 6. These schematic illustrations of the general aspects are supplemented by a l i s t i n g of specific c o ~ t r u c t i o nactivities related t o principal criteria for screening such as type of work, remoteness of site, and level of minimum wages. i n Figure 7. The above guidelines may assist the parties concerned i n focussing on the use of labor-intensive work methods where appropriate, and on programs/projects which would warrant a detailed assessment of a l l rele- vant components prior to deciding on the labor/equipment mix co be applied. Tabla 1 THE STUDY OF TEE SUBSTITUTION OF LABOR AND EQUIPMENT IN CIVIL CONSTRUCTION A BESIURCE AND X I ~ ~ ~ E N T A T IPROJECT O N P R O J E ~COMPLETION BEPORT EONDUBAS: Variation of Coats per Kilometer for a Typical Road in Flat Tarrain a61a Function of tha Volume of Earthwork Quatitias per kilometer of Typical Road Bush Clearing Earthworks Drainage Gravelling (ha) (m3, (1.m.) (m3, 0.9 Variable 30 870 Coats per kilometer - Volume DCMO IBRD USAID IDB DMOCONT (m3, (LPSfkm) (LPSfkm) (LPSfkm) (LPSfkm) (LPSfkm) 1000 21,323 27,485 28,979 35,137 26,893 1200 22,951 28,167 29,455 35,991 28,951 1400 24,579 28,849 29,931 36,845 31,009 1600 26,207 29,631 30,407 37,699 33,067 1800 27,835 30,213 30.883 38,553 35,125 2000 29,463 30.895 31,359 39,407 37,183 2200 31,091 31,577 31,835 40,269 39,241 2400 32,719 32,259 32,311 41,115 41,299 2600 34,347 32,941 32,787 41,969 43,357 2800 35,975 33,623 33,263 42,823 45,415 3000 37,603 34,305 33,739 43,677 47.473 3200 39,231 34,987 34,215 44,531 49,531 3400 40,859 35,669 34,691 45,385 51,589 3600 42,487 36,351 35,167 46,239 53,647 3800 44,115 37,033 35,643 47,093 55,705 4000 45,743 37,715 36,119 47,947 57,763 4200 47,371 38,397 36,595 48,801 59.981 4400 48,999 39,079 37,071 49,655 61,879 4600 50,627 39,761 37,547 50,509 63,937 4800 52,255 40,443 38,023 51,363 65,995 5000 53,883 41,125 38,449 52,217 68,053 Source: Jose Santos Monzon Gamez. "Estudio Comparative de Costos de Construccion Table 2 m STUDYOF TIE SUBSTITUTIONOF OR ANDEQUIPMENT IN CIVIL CONSTRUCTION A BeSEARCH AND IMPLZMENTATION PROJECT Pk0JEm COUPLETION REPORT HONDURAS: Cost Comparfson between Work Methods - Rocurement Arrangements for Cases Inveatiqated 0 Executing Agency OCMO IDB Hypothetical or Sponsor Case Work Work Method labor equipment equipment equipment labor Category -- - Bush Clearance force ace. force ace. contract ) I ) 1 Earthworks force acc. force ace. force ace. ) 1 1 1 Drainage force acc. contract contract ) contract ) contrace 1 1 Gravelling force ace. force ace. contract ) 1 Note:' The equipment used in force account operations was rented. Contracts were based on tendered unit prices. Source: Jose Santor EQnzoa Garnet, Eatudio Comarativo de Costoa de Conattuccion de Caminos en l a Republics de Honduras (Mano de Obra - Equipo, Administracion y Contrato). Guatemala, September 1985. Table 3 THE STUDY OF TEE SUBSTITUTION OF-LABOR AND EQUIPMlNT. I N CIVIL CONSTRUCTION A BESEARCH AND IMPLEMENTATION PXOJECT PROJECT COMPLETION REPORT HONDURbS:' Costs pergtlometer of Road i n Flat Terrain aa a Functionof U n i t Prices .' and Quantities for each Group of R o j e c t s and a Typical b a d Section -- Dnit Prices of each Work Category -.-. - - - - - U/M a 0 (fit.) IBRD(E8t.) USAID (fit.)IDB CMOCONT(Est.9 Bwh Clearance LPS/HA 1,163.12 1,198.62 620.55 1.835.79 1,462.34 Earthworka LPS/U3 8.14 3.41 2.38 4.27 10.29 Drainage LPS/M.L. 122.09 446.39 364.01 403.96 146.76 Gravelling LPS/M3 9.74 11.04 17.38 19.65 12.51 Quantities of Work per Kilometer for each Group of Projects and a Typical Road Section. Bush Clearance Earthworks Drainage (ha) (m "1 (1.m.) m o 0.23 1,750 26 82I IBRD 1.66 4,297 33 976 USAID 0.80 2,877 26 806 I IDB 2.00 5.372 55 826 I Typical Road I Section -0.90 2,000 - 30 - 870 I Costs per Kilometer. Based on Quantities of each Group of R o j e c t s and a Typical Road Section, as a Function of the Unit Prices of each Group of Projects Project *antities DCMO IBRD USAID IDB CMOCOKI (LPrn) (L P ~ ) (m7m) (LPTrn) m' DCMO 25,694 26,967 28,089 34,578 3 2 . & 4 ) IBRD 51,645 43,482 40,964 55,864 65,293 USAID 35,370 31,235 30,787 40,067 44.66) IDB 60.855 54,522 48,514 65,182 76,656 Typical road Section L 30,895 29 463 t L 37,.el 31 359 39 407 Source: Jose Santos Monzon Gamez, Estudio Comparative de Costos de Construccion !a Caminos en la Republics de Honduras (Mano de Obra - Equipo, Addniscraclsn y Contrato). Guatemala, September 1985. Table 4. THE STUDYOF TEE SUBSTITUTION OP LABORAND EQUI~WINT I N CIVIL CONSTRUCTION A kESEABCB' AND IMPLEKENTATION PROJECT PROJECT CdiPLETION REPORT HONDURAS: Costs per Kilometer of Road i n Mountainous Terrain aa a Function of Unit Prices and Quantities for each Group of Rojectr and a Typical Road Seccion -?- C_ Uaic Ricer of each Work Category - UIM - - - - DCMO(Est.) IBRD(Ert.1 USAID(E8t.) IDB CMOCONT (Eet.) Burh Clearance LPSIBA 905.04 1,652.57 1,213.10 1,835.79 1,092.09 Earthworks LPSfEU 10.26- 3.91 5.59 4.27 12.82 Drainage LPS/H.L. 122.89 446.39 337.27 403.96 145.37 Gravelling LPSfM3 12.54 6.24 16.38 19.65 16.43 Quantities of Work per Ktlometer for each Group of Projects and a Typical Road Section. W Bush Clearance Earthwork Drainage . (ha) (n'1 (1.m.) DCMO 0.47 4,031 31 - 675 - IBRD 2.57 7,394 33 1,139 USAID 0.39 1,381 25 825 IDB 2.00 5,372 55 826 Typical Road Section - - - 1.20 4.500 28 - 890 Costs per Kilometer, Based on Quantities of each Group of Projects and a Typical Project Quantities Dm0 IBRD USAID IDB CXOCONT (LPZ~E) (LP~"=) (L P ~ ) (LPTKM) ~~ DCHO 50,027 34,486 44,537 43,770 67,751 IBRD 96,559 55,127 74,331 72,115 121,146 USAID 27,910 22,241 30,058 32,828 35,285 IDB 74,085 54,151 64,639 65,182 92.669 Typical road Section -61,858 L 37 631 50.632 50,217 77,694 Source: Jose Santos ?I laborintensive c i v i l construction tasks. Some examoles of w t e n t i a l user >! the subject. are included in the appendices. ~bchni&l'KAbrind& NO; i 2 ' Haulaqe by Headbaskats, Shoulder Yokes and Other Yanual had-Carrfing Het9oda Octobr 1975 This memorandum describes the use of headbaskets, shoulder yokes and other manual load-earrging methods traditionally used i n c i v i l construction. In addition, relationships for estimating productivity usi3q headbaskets are presented, based on a sinole theoretical w r k cycle "calibrated" by usins the results of production studies currently available. An i l l u s t r a t i v e example is also given showing how the cost of headbasket haulage can be calculated by using the productivity data presented. *;=ha=i ..ee&d& .*;; i3 - me Use of Wheelbartown in C i v i l Construction October 1975 This memorandum describes the characteristics of wheelbarrows, the . - mechanics of their use, various ansects of their desim and features of *eslbarrow mrking, all with particular reference to the task of haulage in civil construction. In addition, relationships for estimating produiztivity wing wheelbarrows are presented. An earliar memorandum (No. 1) dealt with sol. limited experimnts with wheelbarrows. \ Ihis maorandm describes new vork which is complementary to those experiments. fichiicii &otandum.H;.&'i6 R.rdwars Research Summsv October 1975 This memorandum outlines the investigations into the field of hardware made to date. A slnarrmry is given of the scope of future work planned, or needed, in construction hardware research for labor-intensive and intermediate technologies, primarily in connection with tasks related to earthtmrks. ..&& rand*-ho;.i3 The Planning and Control of Productioni Productivity,and Costs in Civil construction Roject Wtober 1975 Zhis memorandup briefly discusses planning and describes a control system which should asgist s i t e managers to make sound decisions based on measuremants and should assist planners to have more reliable data for future projects. The system described is complementary to and compatible with the Fhnual for Roductivity Data Collection, already issued as Technical Memorandum No. 8 of this series. It is specifically designed for use in labor-intensive construction projects. fi~fi*icii~iiemo~anau*.Ho;.ib Lever Cranes October 1975 This memorandrn describes the testing of two tyms of manually-powered lever cranes on a canal excavation s i t e in Lndonesia. The r e w r t shows that i f lever cranes are used for this haulage activity, several factors combine to qeduce the productivity markedly below that for headbasket haulage and concludes that manually-powered lever cranes appear to have l i t t l e application in civil construction, except for a very few special circumstances. .. .. ..... .. .. . ....... .. . Technical Memorandum No; 17 ' Compaction December 1975 This paper r e v i e d the techniques applicable t o compaction o* laborintensive projects. It does not purport to be a treatise on compac- tion, but discussea the on-going processes as noted by observers in various countries and examinas the results of an i n t e n e n t i o n study carried out on a canal construction project in Indonesia where the productivity of various types of simple equipnant, p o ~ r e dby animals or httl~ns,was measured. *i~h~cii'hc~tiriiicirh'ko;'iB Soreadinn Activities i n U v i l ~ b n s t r u c t i o n Spreading is a common but frequently minor activity i n civil con- struction. Observations have been made of on-going methods i n earthworks and pavement construction, including surfacinq, lhis memorandq summarises the results of these obsemations and concludes that for non-bituminous materials there is a relatianship between the productivity (expressed as output per man-hour, etc.) and the layer thickness. It also appears that spreading bituminous materials requires less e f f o r t than that needed for non-bituminous materials of similar thickness. tichiicil'~~oran;tum'*i,;'i9 Excavation February 1976 ' This paper presents excavation data from s i t e s i n India and Indonesia largely from observations of on-going work, but i n some instances intervention techniques, such as pre-ripping by plough, were applied. It is found that i n many cases the excavation activity is inextricably connected with the loadinq activity and that in general la5or-intensive methods are very similar i n cost to, and often somewhat cheaper than, equipment-intensive methods. The use of tracked or wheeled machines for excavation alone is substantially more expensive than manual excavation. . . . . . Technicii ~emoraadumNo. 20 Loadinn and unload in^ Activities Observations have been made of manual loading and unloading of n s t e r i a l s into or from vehicles and other equipment, using hand tools. This memorandm sumnurises the results of these observations. A f u l l s t a t i s t i c a l analysis of the data shows that productivity is significantly affected by the paym&nt method being used and the level of supervision. ~ e c h n i c a lMeiorandum-NO; 21 A Literature Review of the Work Output of Animals with Particular Reference t o t h e i r llse i n Civil Construction. February 1976 This is a summary report of a literature study of the mrk output of animals and the facto.rs which affect animal ~ r o d u ~ t i v i t y .It is intended as a complementary memorandum to Technical ?fenorandurn No. 11 on human erpo- nomics, and should be read i n conjunction with it.. The conclusion is reached that special attention should .ba -. paid to the work of animals h e n uaed on civil conrtruction taska and to articular problem thin work may cause; Qi=htiiiicai.fi...ti.*"m'Eio; .ii Baulage Using Aerial Roueways lbis paper discusses the results of experimantal field studies d t h -- ropeways, and outlines the scope for using aatinl ropewaya i n labor-intensive construction work. It is shorn that in.appropriate circumstances a simple -CC. ropeway installation can give significant reduction in unit cost of haulage compared with manual load carrying. fh~6nicii.&*moiinaigEi~;.*j The Use of Bail System i n Cfvil Construction June 1976 This memorandum describes the methods of assessing the suitability of rail systems for haulage m r k i n c i v i l construction and describes various types of systems used experimentally for such work i n India. t i ~ h r i i c a i ~ ~ ~ m o i i * ; l u m m ~ O ~ ~ i B The Use of Agricultural hactor/Trailer Combinations This mmoranduu describes the use of agricultural tractors t o haul construction materials in trailers, explains the problems i n their use and discusses ways of overcoming these problems. Various options are analyzed, such ae the choice between two- and four-wheeled trailers, tipping trailers and ballasting of the tractor. A large portion of the memorandm is devoted to the question of load transfer from t r a i l e r to tractor and to the manner i n which this affects the traction available to the tractor. The paper gives a numerical guide to the selection of appropriate combinations of tractor and t r a i l e r size for a given haul route surface and gradient, followed by an ex- planation of how to calculate the numbers of laborers and trailers required. Detailed instructions for calculating productivity and unit costs are given, and the relative merits, i n cost terms, of truck and tractor/trailer haulage are discussed. The memorandm ends with a review of other implements avail- able for use in conjunction with tractors, such aa rippers, graders and rollers. fi~hiicif-ficmoiinduih.iso;~i5 Aggregate Production June 1976 This memorandm discusses aggregate production with particular reference to labor-intensive and semi-labor-intensive methods, and gives the results of field observations of productivity. As a result of these observa- tions it is concluded that for handbreaking of stone the size, shape and con- struction of the hammer is important, and more attention needs t o be given t o the steel used for the hammer head. In addition. there is a relationship between manpower required for manual breaking and the size of product, reduc- tion factors, and hardness of the rock being processed. fi=&i=ii