WORLD BANK A METHOD FOR CONSIDERING POSSIBLE ALTERNATIVES TO THE ORGANIZATION OF RURAL ROAD MAINTENANCE AND ITS APPLICATION TO A SUB-SAHARAN AFRICAN CONTEXT June 1982 SCOPy Melody Mason, Consultant Projects Advisory Staff (PAS) Project Cycle Unit Technical Note No. 1 This Note is one of a series issued by the World Bank's Projects Advisory Staff to assist staff in the Bank in the cross-sectoral aspects of project preparation and execution in developing coun- tries. The views and interpretations in this document are those of the author, and should not be attributed to the World Bank, to its affiliated organizations, or to any individual acting in their be- half. ABSTRACT Classical approaches to improving road maintenance organization in Africa have largely failed, mostly because they did not deal with the large number of constraints and pressures external to the organization itself. The following steps, based on the framework outlined in Staff Working Paper No. 375, are therefore proposed for designing rural road maintenance organizations under the conditions prevailing in a typical Sub-Saharan African country: (a) make the external environment more favorable to the organization (e.g., identify the -road beneficiaries and enlist their political and/ or financial support); (b) ensure better performance by the other orga- nizations upon which rural road maintenance depends, or minimize such dependency (e.g., use alternative technologies such as the "lengthman" for routine maintenance); (c) improve motivation within the maintenance organization itself by using locally proved incentives, better relating training to the field reality, removing bureaucratic bottle- necks, or developing a local contracting industry to which certain activities will be entrusted; and (d) ensure that the lessons learned from r6ad maintenance are taken into account in the design of new roads. The paper emphasizes that there is no one solution for improving road maintenance organization, and recommends that designers and man- agers consider which among the many possible options best fit the pre- vailing circumstances. Prepared by: Melody Mason (Consultant, PAS) Copyright tC 1982 The World Bank 1818 H Street, N.W. Washington, D.C. 20433, U.S.A. GLOSSARY LIFP: Labor-Intensive Feeder Roads Program MO : Maintenance Organization MP : Member of Parliament MT : Ministry of Transport MD Mechanical Department PE : Provincial Engineer PME : Provincial Mechanical Engineer PRE : Provincial Road Engineer TD : Training Department A METHOD FOR CONSIDERING POSSIBLE ALTERNATIVES TO THE ORGANIZATION OF RURAL ROAD MAINTENANCE AND ITS APPLICATION TO A SUB-SAHARAN AFRICAN CONTEXT TABLE OF CONTENTS Page No. SUMMARY i I. INTRODUCTION 1 II. ROAD MAINTENANCE ORGANIZATION: THE BACKGROUND 2 Road Maintenance Activities 2 The Zandya Road Maintenance Organization 3 Effectiveness of Road Maintenance in Zandya 4 III. IMPROVING THE DESIGN OF THE ROAD MAINTENANCE ORGANIZATION 7 A. Deficiencies of Traditional Approaches to Organiza- tion Design 7 B. Alternative Measures to Improve the Design of the Road Maintenance Organization 10 Long-term Policy 11 Strategy and Structure 12 Political Environment 12 Centralization or Decentralization of Feeder Road Maintenance 13 The Financial Environment 16 Improving the "Controlled" Environment: Motivation and Incentives 18 Supervisory Staff 21 Choosing Alternative Structures 22 (a) Contract Maintenance 22 (b) Casual Labor 23 Choosing an Alternative Technology with Differ- ent Organizational Implications: Capital vs. Labor-Intensive Maintenance 24 The Integration of Development and Maintenance 25 Administrative Environment 26 IV. CONCLUSION 27 LIST OF CHARTS AND ANNEX Chart 1: ORGANIZATION OF THE MINISTRY OF TRANSPORT (MT) WITH RESPECT TO ROADS Chart 2: STRUCTURE OF THE MAINTENANCE AND MECHANICAL ORGANIZATIONS Annex: ANALYTICAL FRAMEWORK FOR ORGANIZATION AND MANAGEMENT OF DEVELOPMENT PROJECTS PREFACE For a long time, the World Bank has been concerned with finding ways to better understand what lay behind successful organization of development projects and with developing methodologies which would help both project designers and project managers. Staff Working Paper No. 375 on the Design of Organizations for Rural Development Projects was intended to start filling the gap by providing a framework allowing further experimentation. The widespread interest that the paper generated, and the overwhelming support it received among both academic circles and practitioners, in developed as well as in developing countries, convinced us that we should indeed experiment further with the framework. This was done by testing it against the reality of one of the most intractable organizational prob- lems: the maintenance of rural roads in Africa. The results of these efforts are summarized in this PAS Technical Note which was written by a former colleague of ours, Melody Mason, a transportation economist, sociologist and organizational specialist. Briefly stated, the design task consists in finding ways to organize road maintenance activities under the following conditions: (a) the concerned Ministry's recurrent budget is limited and unlikely to increase in the future; (b) equipment operates about one-third of the time and endless delays are encountered in procuring spare parts; (c) political interference is unavoidable and results in continuous requests for diversion of maintenance equipment for non-maintenance activities; (d) the large permanent work force is subject to civil service regulations affecting the hiring and firing of staff as well as reward systems; (e) the work force is poorly motivated and weakly super- vised at the supervisory and engineering levels; (f) a variety of donor agencies each having their own standards, approaches and priorities, is involved in the road sector; (g) government shows a limited concern for road mainte- nance. Melody Mason suggests that there are a number of options open to designers and managers to deal with each one of these factors though none alone can be expected to provide "the solution" to improving road maintenance organization. - 2 - To put her findings in a nutshell, if one detfrmines, as all coo often occurs in this sector, that neither intended beneficiaries nor Road Agency staff are really interested in road Tai.nance, it will not help much to limit the organizational effort to specifying which tasks should be done when and by whom, the reporting ralationsihip, the budget procedures and the like. Instead, organization design of r)ad mainte- nance also involves: (a) ensuring that the environment over wtich te zation has neither control nor influence be -a more favorable to its purposes (e.g., by -:akd; ca and beneficiaries better understand the :ecss: t maintenance, involving them in programm.in at-" nance work and in enlisting politi-al and Fi7ancial support); (b) ensuring that other organizations u on wich road maintenance depends (e.g., for mair:teane m;ipment) will respond better, or, finding ways (e.!, through use of alternative technologies) to ministe such dependency; (c) improving staff and managerial motivaticr zW%rough methods particularly appropriate to -he >:1 environment, or by contracting out certal acti'.- ties. The options proposed to improve the external * internal environments of the road maintenance organization wre :erived from observation of the daily activities of field engineer,. discussions with Provincial Engineers and senior government offlcIals -ostly in Sub- Saharan Africa, as well as from the experience of con.zing firms and that of our colleagues in the Highways and Rural De ----10aent Sectors in the World Bank. Their encouragements and contrLbut...- are gratefully acknowledged here. In conclusion, it is our hope that thi3 oapez will help !ihway designers, public works managers and other deci3ice wifers in t.e Trans- portation Ministries of our member countries bec-er 4i.-nose or:anima- tional questions concerning road maintenance and ens:.- age the ~'.abo- ration of engineers, socio-economists and organizati.. Jtzsign spia- ists in devising the imaginative approaches that tlhtr iolution requires. Francjs -ethem Technical Cooperation Adviser Projects Ahdviscry Staff A METHOD FOR CONSIDERING POSSIBLE ALTERNATIVES TO THE ORGANIZATION OF RURAL ROAD MAINTENANCE AND ITS APPLICATION TO A SUB-SAHARAN AFRICAN CONTEXT SUMMARY i. Increasing attention has been given during the past few years to problems of road maintenance in Africa and other parts of the world, but increases in the standards of maintenance have not been easy to achieve, despite the resources devoted to such efforts. This paper has two objectives. First, it attempts to understand the rea- sons for the difficulties encountered in road maintenance in many developing countries. Second, it proposes a range of alternatives to help organize the maintenance of low-class or feeder roads, ranging from earth to semiengineered roads, since these roads generally form the major part of any road network in developing countries. Engi- neered gravel roads are also included where they serve essentially as feeder roads as opposed to forming part of the main road network. The paper is based on extensive discussions with field engineers responsible for road maintenance, mostly in English speaking African countries, and a review of their managerial and organizational initi- atives. To focus more clearly on issues, the road maintenance organ- ization of a hypothetical country -- a composite of the author's experience -- has been used as a case study. ii. The paper originated from the disappointing results in many countries of comprehensive efforts to improve road maintenance organ- izations, even when funding was increased. This suggested the need to examine whether there were questions other than financial ones, that affected the ability of maintenance organizations to improve output markedly. Among the factors being considered are organiza- tional weaknesses as well as diversion of resources to other nonmain- tenance activities. Organizational weaknesses, in the Sub-Saharan African context as elsewhere, are commonly attributed to poor design of the organization, inadequate procedures and channels of communica- tion, insufficient and inexperienced trained staff (particularly supervisors and engineers), and cumbersome bureaucratic procedures, especially with respect to procurement. The organizational weaknes- ses are compounded by financial constraints, restricting the purchase of replacement equipment, essential stocks of spare parts, materials, and so on. The response of most governments to these problems, gen- erally with external assistance, has been to redesign the organiza- tion and increase training; the consultants hired to carry out orga- nization studies have tended to concentrate on structure, reporting and cost control systems, numbers and qualifications of personnel required at each level and other resource requirements. iii. The problem with this approach has been two-fold. First, the concentration on the role of management in dealing with questions internal to the organization has meant that external elements that can affect organizational performance have often been ignored. As - ii - outlined in Staff Working Paper No. 3751 and in the Annex, it is es- sential for designers of organizations to understand the environment in which the organization has to function. To what extent can a man- ager be said to control maintenance activities when he is subject to direct interference or irresistible political pressure to devote main- tenance resources to particular nonpriority roads and nonmaintenance activities, or when he is subject to government procedures which limit his ability to spend his budget on necessary resources and lead to lengthy delays in maintenance operations? To what extent can a main- tenance manager be effective when he is faced with a public who has little perception of the importance of routine maintenance and of pre- serving something for the future by taking appropriate measures today? The impact of such external factors can be such that effort to improve organizational performance through changes in structure can be completely neutralized. iv. The second problem has been that the emphasis given to the pri- marily inward-looking management role has nevertheless often over- looked such crucial factors as management style and the level of staff commitment and motivation (that is, the "people problems") within the maintenance organization. The effect of the civil service structure on organizational performance (for example, the effect of low pay, job security, and bureaucratic promotion systems on staff motivation), the low level of cost-consciouness among staff, and the emphasis by soci- ety in general on access to, and distribution of, resources, as opposed to efficient use of resources, normally has been underesti- mated or ignored. Western concepts of management have often been rec- ommended without realizing that their actual adoption required the maintenance staffs to share the designers' objectives, that is, to reduce costs, increase efficiency, and so on. Since these concepts are often new to developing countries, it might be more beneficial first to devise ways of making people want to be efficient and reduce costs. v. This paper therefore proposes that more analysis of the organi- zation and maintenance activities be carried out from the perspective of the managers at the successive levels of the organization. This will mean looking at the environment in which the organization has to function, both in terms of how external entities can aid or obstruct maintenance activities and how the environment can affect the internal management of the organization. The latter will mean consideration of such constraints as the civil service structure and procedures and the cultural attitudes and values of maintenance personnel. The final design of the organization should, therefore, minimize the obstacles by directly avoiding certain problems or by building in coordination between the maintenance organization and external entities. The 1/ William E. Smith, Francis J. Lethem, Ben A. Thoolen, "The Design of Organizations for Rural Development Projects: A Progress Report," World Bank Staff Working Paper No. 375, March 1980. - iii - design of the organization should also capitalize on the potential support in the environment and on existing organizational strength both within and without the maintenance organization. vi. More concretely, the paper recommends consideration of the fol- lowing possible organizational alternatives: A. Increasing involvement of local administrative levels in program- ming maintenance work. The purpose of this recommendation is: (a) to make the local community more aware of rou- tine and periodic maintenance work; (b) to let the local community have more say in the relative priorities of roads to be main- tained and at what level; (c) to make the local community more aware of the limited resources available and occasionally to elicit contributions from them for specific work; (d) to curb the pressure on the Maintenance Orga- nization (MO) to divert maintenance resources to other uses, especially at the local level, and to channel prominent local individuals' requests for work on specific roads through local committees. There are various ways to increase involvement at the local level and those involved may range from District or regional Councils or their equivalent to special road user committees of local representatives. The local councils or committees could then participate with the MO in programming maintenance work, with the MO reserving a certain amount of limited resources for priority routine and priority periodic main- tenance activities. The MO should explain to the council or com- mittee the necessity for these activities and allow them to select other priorities. If the council or committee decides that certain unclassified roads should be maintained because of their local import- ance, then the MO could accept their decision (with some possible exceptions) as long as the council or committee were able to say which classified roads not to maintain. This would help instill a sense of cost-consciouness and encourage the local population to think in terms of relative priorities and tradeoffs. Local representatives could be informed when funds, transport, or equipment were not available and could put more pressure on Government to obtain the resources. Alter- natively, they could try to raise them locally. These resources would include, for example, funds for a spare part for a grader or truck, funds for hiring casual labor (or contributing labor itself), or the loan of transport from the cooperative or other source. - iv - B. Raising more funds for road maintenance. A niviber of alternatives exist such as: (a) automatically allocating a cer:tin perzentage of road user charges to maintenance (unlikely in many afr4ican ccuqtries' present economic situation, but perhaps something i. ainmr; (b) financing maintenance activities through cons t :Iuctn rc while building the ministry's or local contractor ' zaiae capac- ity; (c) including more maintenance activities i a oo programs; (d) donor financing of spare parts; and K a n of vehicles and equipment by commercial agencies. C. Experimenting with methods for individual man : ;.se staff incentives. One of the basic problems of f ork is staff motivation. Thus, far more attention has - to :!otiva- tion problems, for example, through imaginative a e and time off and rearrangement of work into team wo,-, asks," and so on. Whole tasks require more control of r the maintenance and repair equipment. One possibili , there- fore, is providing equipment units with their ow: fully equipped as possible), and giving specific tasks 7 f 4 - incen- tives for meeting targets. This approach would . level of duplication and underutilization of mechanicav `-,t the costs of duplication have to be balanced against of inde- pendent units (just how independent would have t. e by trial projects). The same argument could apply to the prov .ys At present, the typical maintenance organization . tral workshop for overhauls, one regional workst- 2, repairs, and district or local workshops with or'' f-r simple repairs. Closer attention needs to be g'.,.--- facilities successful contractors have for the . t their fleet, since they face the same trade-off of -c ost of underutilized workshop equipment and other resoti- ,a co':st of waiting for repairs to be carried out at a disti Since government organizations are not as efficient a t, r-racrors, it may well be less costly for the maintenance a,i-e better- equipped district or local workshops. One of t * r the high costs of the present system is the limited capac- in the maintenance organization to program and implemer - task of bringing equipment from different units into wa - i-naired, and once there, programming and implementing pr. - -d returning equipment to units. D. Examining ways to improve the performance : -e- cially at the lower level. One of the problems er visors are taught too much in one training sess -:ted to perform as highly in the real world, with all . as n the classroom. Supervisors may first need to te at a very low, clearly defined level, such as .x onl; on culvert, ditch, and bush clearing, before p: e COM- plex activities such as patching of paved roaA-- a id - v - resealing. This would be in line with task specialization mentioned above. The same principle of progressive training could apply to mechanics and inspectors. In addition, follow-up is necessary by the Training Department (TD) of recently trained candidates in the field, with emphasis on programming work. E. Developing local contractors for routine maintenance (as is being done quite successfully in other countries, including the USA). An argument often cited against developing local contractors is the high failure rate, but this needs to be balanced against the failure of many of the attempts to improve the efficiency of the Mechanical and Maintenance Organizations. By developing local contractors, the main- tenance organization relieves some of its scarce supervisor resources of activities such as procurement and delivery of supplies, organizing staff and equipment in the field, and motivating staff; in addition, it relieves the organization of some of the burden of procuring and maintaining equipment. Of course, the organization will still need to supervise contracts to ensure that works are carried out according to. the terms agreed. In many African countries the relative lack of sophistication of local contractors makes it possible for the mainte- nance organization to exert a strong influence over them (as part of the training experience) and to control prices (normally unit prices are fixed by the maintenance organization in agreement with contrac- tors who do not have the experience to cost out and submit bids). F. Other possible organizational improvements would include: (a) more flexibility in choice of technology (for example, labor-intensive instead of capital- intensive) with different organizational implications, for example, the "cantonnier" or "lengthman" system; (b) consideration of which Government procedures (such as procurement procedures) might be streamlined, given the considerable impact they can have on every Government agency's performance; and (c) greater use of casual labor to give more flex- ibility to the organization, reduce the staff motivation problem, and spread the benefits of employment to a greater number of people rather than to only the present permanent employees. vii. In discussing the above possible organizational improvements, the word "alternative" rather than "solution" is used purposely since there is no one solution or prescription to improving road maintenance organization. Rather, there are a series of options, some or all of which may help senior officials and maintenance field staff perform - vi - their work betrer.- Many of these have 'Seen tried outr or suggested by Field Engineers, Provilial Engnmers and her sentor staff with whom the at or coll,aboraced diuring In-4t ' 1e years. 1f the reaction to thts !paper is -o mnd.erstand the noM, but : disagree with the altrives ge favor o rnatvs, then one of ics ob es would hre bee achied,hat_, to pave -be way for Ch4S klnäJ o5 discu,-44or.. TN- time funtadsuso sripe,sic ma:t, professIonais 11 thLs fiel1d, IN zuntrls, on the Bank's staff, as well as wi consulzing - aen ecig incresingly aware tat the scain4atä prescr*kptý or v : o st ts of ma.intnce oganizatss are not . a :e fundamtal viii.. `he pro3posals pu forward il m pupt-, 'od ts< thaýt häave alread been tried ad :ose that d . t p a iot basii eae based on de -sumption rdr.:r awl ; h co-trty :o maintal lt s that tC :,:ua. _-'s, rs.o oo iniLia, and that aement re41 1 > , impr e performanc, Ax decrease t i par- ticlar -ountry wh approach iI s not thera, -r the managern are really w ni D : then the alternirives suggcesred to-r reduciaUg -,=a e ce esadiprov- ing incte nal effi e š+ey oill not bc - '-,c ''ear ced tc,nsider easures ' contrac .::ät and tr --,ke adequate t-s: for t estv rsres be;arse of poli ti'.g, ssure. - emotn point is9 ct to uch a g Zati1 iign stage, to r a ctnization accr. , ix. "e paper 41 ured w The . r:uarizes the m ar reports to the Chief Ieinee-r - tne Roaca Department (Chart 1). 2.08 as for many maintenamce organi atitr :hreughout the world, the province has been made the basic uit :-f :gent for both the MD and MO, with each province having ahing from 3,000 km to 15,000 km of low class roads to mairarin.. Th Povaai Engineer (PE), who reports to che Engineer-ir,-,'hi f of e Mi istry and has been charged with co-ord'sting road xcItLes 1: the rv,rince, is the ministry's authorizine. 'gent for the r-once, ald 1 -&d financially resposi- ble for all ..penditure' n yads, i1iudic `unds ,allocated to the MO ani, 5, an- - special pr 'cos, for a in the area of feeder roads. 2.09 Char-e.ristic of de classic '- acon, there is a clcenrly defined and '-. erarchica ! magerial sy,ý, -! -: a pyramid type of structure (Chart 2). The smallest un a road camp headed by an overseer or foreman: ot road camp -A rt average of apprz:i- mateLy 150 km of roads to %aintain. arture from the clas- sic c,7pe of ar3anizatir is at the Pr- 71-ad Engineer (PRE' and Provincial Yec'anical '--. leer (PM 1 e they have a dual reporting rol' to botc PE and ý etive Chief Mair to- nance and Chief Mechan4; Engineers ,ntry's capital ci:y. This gives the PE the au-. -rity he rr the provincial level, although he is still ~ -.ent on - 7 1,-cated from Headquar- ters. The M häas a e 1 or;an4 . .ruxzture (Chart 2): there is one large ?rovinciL'; .rkshp of cach province, plus district w~ shops far t -ying c-, r rs. The central work- shop in the country's .- al ci e overhauls and major repairs whi nor-all, _.raot ze * the province. 2.D The T7 is responz_e for at all levls up to senior surz-rintendent frx Soth hose trained include dri-ers, o,-ators, mec¾jís, - a en, and inspectors. TraIning cirri-d out - the sport school, except for aspe - and superzn -tt r. polytechnic courses under the -.istry of Pd-ri1 al has its own trainin- brigade - rehabilita a Tralining School araa. The TD progrums :rainig ces Mini sr's operatio-.al depart- ments acQ to C1-,ir Ef f ectiv af Road Ma- -:tr- 2.11 11 the condi i ds maintained 'y -.e MO varies fron one '-'iace to axnut. ýJ. 1. level of maintnance throughoi country stil l to be desired. Availabilitv - 5 - of vehicles and equipment is reasonably high and includes a large num- ber of new vehicles and a large stock of grader spare parts. Training has been carried out for a number of years and more than 4,000 mechan- ics, drivers, overseers, foremen, inspectors, and superintendents have attended courses in the last ten years (including repeater and upgrad- ing courses). The maintenance organization has also benefited from strong leadership by the ministry, not to mention some very able and dedicated PEs who have been able to achieve remarkable results in sometimes difficult circumstances. Nevertheless, Zandya's earth and gravel roads still have numerous ruts, potholes, and high surface roughness levels (resulting in higher vehicle-operating costs). This is in part due to poor road maintenance, although other factors such as inadequate drainage, overloading, and higher than projected traffic levels have also contributed to road deterioration. Water penetration into the pavement structure has frequently occurred because culverts, drains, and drainage ditches have not been cleared, cracks and pot- holes have not been patched, shoulders have not been leveled (pre- venting run-off and evaporation of water), and roads have not been re- graveled. Some gravel roads are in need of complete reconstruction because they have not been graded and regraveled on time, and some earth roads have reverted back to bush because they have never been maintained. 2.12 A part of the problem with the road maintenance organization has been MO's dependence on vehicles and equipment. It is not unusual to find that maintenance vehicles and equipment are in working order only 60% of working time (normally taken as 200 days per year) and then utilized for maintenance work perhaps only 60% of the time avail- able. This means equipment is being used only about 36% of total work time (admittedly, in some part of the world, it is even less). To compound the problem in some cases productivity of equipment (such as graders) while in use is also low. 2.13 The availability of vehicles and equipment has been low in the past, but some measures have been taken recently to bring the overall availability rate up to 60% (although doubts have been expressed about the reliability of this estimate). Low availability of equipment it- self accounts for part of low utilization rates, since the breakdown of a few items of equipment often means that other equipment lies idle (for example, tippers for regraveling units). The reasons for availa- bility rates not being higher in Zandya are: (a) the lack of funds for, and availability of, spare parts, but with external aid a stock of spares is now being built up; (b) a substantial portion of the equipment fleet is beyond its economic life, but again, external aid is financing purchase of new equipment; (c) lack of preventive mainte- nance; (d) poor quality of maintenance work, partly because of insuf- ficient numbers and inexperienced mechanics, although increased efforts are now being made to train far more mechanics; (e) operator abuse of maintenance and maintenance equipment; (f) inadequate work- shop facilities -- both buildings and equipment; and (g) a wide range -6- of brands of equipment, which requires a mechanic to familiarize hi- self with many brands of the same veh cle or equipment. The problem o staffing mechanical departments is partic-ularly acute in 'andya, si.nce many :rained scaff leave for e ,rivate sector because the demand for ?tchanical staff is so i and because che privraCe secter affers higher benefits. 14 Another reason for pýur maatec is low prduczIity of aor and %gaipment, even when r,e e sich as equipment, labor, and ae are avaiable (para. 2d17. Lo pr dctivity has been due zo *s: factors, one of which las been h hrage ani ualty of te nvyrvisory staff. The conry has e grat strd&s in trainin or supervisors, gl:hough getting w, >urkvisors o ao their r £.de ~ately has ofter, been a real proT:-z,I is o aien loing 2_ *:outine maintenance, for ,,stance, ht the road ta , are n:t he- aiell supervistzd, nor is theirwor. e routireL iaected 'rsasequently, there is little contro:l wýin MO (a-dou :his d:es ear apply to all provinces), resultir. : porting sys breaking 4n or containinýg unreliable nformavti'. Because "- d poor cal- z of supervision, it is di,ficult o to whac t there is a teal shortage of supervisors s-ince if :Y roductivitr ,;spervisors is low, more staff are req»ired to dc rh nne amount a k Never- 'heless, the ratio of one over3eer to e re an twentý storrs is -:ite high, although sometimes they i svsted by oro head- 2 13 One of the majcr problems is th - rvisory a a the ^perational level are not programming w ,-rk effecti.ly. Apart rm the overall programming of daily w hiy or dering supplies (whether fuel, gravel .-» . en, or , vings) and rzanging for the crs o of labor ard ision of eq _rgent for sriodic maincenance uits. requires com ' le organia v: onal capa- it, whic, 's s':Ua scarce resolur: and,ya. .- prgramming zans hat rizch tita 2s he waiting for is or and sent to arrive, and4aehftoain are ca twithb.l-nopi m . -source.- 216 Cumberoeme burzauczatic przcedure3 vc help r aprove eff- c-J-ncy. The 1Xntralý Tndeýrs- Board proce;é' Vs tten : and the procurement rf spare parts vehicles, ear v:, and gran - zan take mýor:ths, even when the pross goes smoot ,otuna.- v ew man- ag_rs in maintenance organ.aations take s of these »thy delays in planning their wzzk. Up until r : , most Jak e parts ne been purchased only -aseeded, and s:= esay have -.-:me from road and local dealers ae often un-able o knany s because strict im-port quotas, dere have beer s months' vay in deLivery. irsistence on accpepting the lowest cid often reaits in the ing of a local supolier - mavufaccurer who is unable ::e et the 7-,uirements of the -ontract on chedule, because of his lizie is:;acity. The PEs tcmpah it-Wt the amount of paper wcr, to be pr:- sed, and the lenschy tira - L.s for the Procurement artmenc - 7 - to fill requisitions. Other factors, such as releasing the mainte- nance budget only three months at a time, have caused the MO real problems since more funds are required at special times of the year. The argument for releasing funds on a three-month schedule is that if MO received all the budget at once, it would spend it all in the first few months. 2.17 Last, but by no means least, has been the problem of insuffi- cient funds for maintenance, which was particularly bad during the last few years. Frequently, work was stopped in the third month of each quarter because funds only for the payment of wages were left. Thus, during this period wages were paid for very little work since there was no money for the hire of equipment or fuel to run it. Also, MD was unable to purchase new vehicles and equipment to replace un- economic items, which necessitated higher levels of-expenditures on repairs beause of more frequent breakdowns. The road camps have been particularly affected, since they depend on transport to get to the roads. More often than not they have been without vehicles, and although laborers will sometimes walk to roads up to 5 to 6 km from the camp, a substantial portion of routine maintenance has been neglected. 2.18 Another problem which disrupts work programs is the diversion of resources from the maintenance organization. Vehicles and equip- ment are often commandeered for famine relief, for distribution of basic foodstuffs throughout the country, for transport of water, and for road construction and other projects. Funds allocated for mainte- nance are often spent on new road construction or improvement, usually seconding the best skilled manpower available, at the expense of peri- odic and routine maintenance programs. In any one year, this diver- sion of funds can be quite substantial. It is to the credit of the maintenance organization that it is able to respond to certain needs, especially famine relief and maize distribution, when other entities in the country cannot. III. IMPROVING THE DESIGN OF THE ROAD MAINTENANCE ORGANIZATION A. Deficiencies of Traditional Approaches to Organization Design 3.01 Despite the financial and technical assistance received from bilateral and international aid agencies since the early 1970s, the performance of Zandya's maintenance organization is still disappoint- ing, with low levels of productivity, both for equipment and labor. Inadequate financial resources have certainly constrained activities, although during the last two years the MO has received a large number of vehicles and equipment for which it did not pay and which do not come under the hire charge system (they belong to MO, not MD). - 3 - The condition of the roads does not reflect the $900 allocated for each kilometer of road. Therefore, of more concern than the financial issue is the adequacy of maintenance carrie; out witl the reiources available, which immediately raises the isue of dhther t ia!ie -hould be put on allocating more funds Lo cke iaintiance ganization -r het:!er a major effort should firs: le nadi t: Improvr. c if axisting resources. 3.02 The recognition of perzrfmance groblms ) rf >;aniza- tional weakness usually Iads ta organIzati --e which tend to assume that all the actntie .ecessary to a- -..tenance car- gets can be controlled by t-e rganiza-ion ro!ment external to the mainter-,:e aa-cy can be is sided.3 1s a result, maintenance orc1iatios tend to 1--o l 2 ed, and errors of design tend to be d's As problems nor implemen- tation. But, in reali 11e !nager In t 2Ynstantly deal- ing with factors beyou 1ol, ar o% * - litt]e influ- ence. Since tradition.. aization r ntrated only on the role of management a ! nored thf "_:;-,1ents that can affect organizational and ovev r little or no control (such as exces .Leal press interference, civil service salary l bureaucr' ¾, the mainte- nance organization's _`improve o, ptformance through changes in stt. oe comp 3.03 Even with the a the prima < ing manage- ment role, insufficie.: -i has bee ve 7 ultiplicity of nonfinancial incentivr . 1.29-3.3' availiZ -nagars to increase productivity. `iv , it has b-een ass-- ac by designing in detail the structur. - organi:-: as i o oeraional systems, by providing trai:i.. ' and by p 1ing hcw -an y positions there should :h 16 organ i n will func-ion well and require no ot"- -ati Aside One o the reasons for strucr ig 3atent L that it avoiis the "people problems,' :tion zemen i4, and the pr;b- lems of commitment and a. of the :ac:ors in how effectively any one p- ' Zandy is the ::age'fr. We and creativity of the PE a enior :-sory sey The .:ruc:ure of the organization has 11m :eet a ch manager trenghs, 3.04 One example of '- vssas arE intraduced 1w :r to imrove organizational efficiecy in maintenanct. organizat-.- is th± hire- charge system for the equtnme-nt fleet. in Zandya, b hire-char3e sYs- tem has not resulted as a ='t savings, partly becva- :f the of cost-consciousness ang ea maintenance staff an PrtIY beci-se supervisors are not tryLng to t -1!e naxium use o of equipment o that money saved on hre Cmarg&as c&a be sent on hiring more eipment, 3/ Staff Working Paper ?N. '!5 op -ci 4/ R. Waterman, T. Fiters, c;a'd J. Phillips, "Structure is not Organiza- tion" (Business Rizr-, *:Ia 1980). - 9 - on fuel, on materials, and so on, even in times of severe financial constraints. One of the reasons for this lack of efficiency is that if money runs out the staff cannot continue working but still are paid. If work does not get done there has not been sufficient accountability (as in many civil service structures) and explanation about funds running out are too easily accepted. Widespread adherence to western concepts of productivity, efficiency, and individual achievement may be found in a number of industrialized countries but may be quite new concepts at the supervision level in a number of African countries.5 In the rural areas, values such as access to, and distribution of, resources6 and social interaction ("affilia- tion") may be more important.7 Access to, and distribution of, resources means here the priority given by society to making resources available to the community, such as education, health services, and roads and ensuring that within the community such resources are seen as being fairly distributed. For example, studies have shown that many cooperatives give more priority to obtaining resources and then assuring their fair distribution than to concentrating on the most efficient use of those resources. It is more important to them that resources are there and are seen as being fairly distributed. 3.05 The effect of civil service structure on organizational perfor- mance is generally not given sufficient attention in organizational studies. Civil servants are low paid compared with the private sector (except for unskilled labor) and bonus systems are not allowed. Financial incentives to staff are therefore not possible. The lack of alternative incentives in the maintenance organization stifles initia- tive and motivation and results in a sense of inertia. Whether one performs well or poorly often goes unnoticed. Promotions are limited at the lower levels and are usually based on educational qualifica- tions not always relevant to the added responsibilities and not pos- sessed by a large part of the labor force. The good people at the supervisory level often switch to other, more successful programs within the Roads Department. The job security traditionally asso- ciated with civil service employment encourages transfer rather than dismissal of poor supervisors, with the result that problems are merely shifted to another part of the organization. Finally, admin- istrative systems have had difficulty in learning from past mistakes. For example, too often good management may be exercised by one excel- lent PE, but when he moves on, his managerial expertise is lost because the management learning process is not institutionalized. 5/ J. Moris, "Transferability of Western Management Concepts and Pro- grams: an E. African Perspective. (Rockefeller Foundation, Educa- tion and Training for Public Sector Management in the Developing Countries; March 1977.) 6/ D. L. Leonard, "Reaching the Peasant Farmer: Organization Theory and Practice in Kenya" (University of Chicago, 1977); and G. Hyden, "Efficiency versus Distribution in E. African Cooperatives: A Study in Organizational Conflicts. 7/ C. C. Onyemelekwu, "Men and Management in Contemporary Africa," Nigerian Institute of Management. - 10 - .ndeed, a large prvices even good PEs have difficulty mai;ntainlig conr,tro! eincå the orgsnisation becomes too large and complex for ona *.c,c uiv to mattage. 3 { Anc` - externai l :in affects tne performance of the ain- terance cryaizations, is te r ereral pbl's attitudes t- an per- cmaI nc. Th13 is a jidespreadt probl3, een ia Kore developed countrile. t e such as cleaning of c drans, and ditches do act eem s necsay to the general qopuation because they make no nt.' t difrence to the look of the road zts,tf, and the prever`re t areat realizcd. Nor is aring Sdrans tnsidered -rr~ by the s cal pub`z despite the dis- :: ~:fect of blocked ±ins -n t-'e rad. Yet when roads do m,t d yeteriorats bctause o uater 5-oratin there is a public J . ecause the publi< confuses & ies, ch as graveling an road and reh'abili arion r maiitenance, there is uttch more pressu71e for the i tan for many of the uatenance activi,.es. In edt;cion programs to n :le J public 1m the 'ural areu t e need for road mainte- `ce my he less saczessfr in subis c cultus, since people's Srizon my be more limited, unless s )cogrars are combined with ,casureq r,. give the commuzrty some cnnctnl of, and participation in, road mair!nace to efncource thkir 11nreneent and commitment. 07 ?< s of tZ ,robi.t :as : he tritional approach to eganiza.-i design, this ti: at mr rmalysis of the ganiz ~ of maintenance arried :ut from the ,er- pective :he extar-:,al eo.-, whiz1 cne manager must .rk. will require the nza nalinkages with stitut_< ~ and ind-viduals 'on ý zrucial for achiev- - the n etives of the S' a nIzat1 c, Such a design 3t r ..z e that th.e par- .- :e in--l,utions and indi- ~uals aways partial M ind i. .ssentially it t :cess. ;ays a u ý"e fo"- to indtce the`r rt ic-: in;, namelyv, crea :js his ý:tý-;t1 me et t '1r own Altrnat-ive Meauvres t gn cf t`Äe Road Mari tenance æni:aton 3 This section of the _ s or various alt,.a,- ives - to Tanv frican xnunt ad ra4ntenan-ice ori_`za- i. The word "alternati lut n" is used ;2posely e t.,ee inr:o one solu ( cosltant sa:dics _en pro*-se, to improvin Z orga-izat-:n. Ri-t, 'e are a 4erias of opt whith 1.- help Lfrian ernments, senior ofr:i a na field staff perf_r= -eir work '>etter, if thav These glternative-3 need be applied on a pi13t basis to me nica paricular set of circum- stances they are most effecc!e number of these alternatives are actually being tried ont Ir n -Saaran Africa, were being considered - 11 - by field engineers, Provincial Engineers and other senior government staff with whom discussions were held, or were inspired from discus- sions with these officials. Long-term Policy 3.09 When designing maintenance organizations, long-term policy and objectives need to be determined, not only by maintenance agencies but also by all those with a vested interest in road maintenance (the "stakeholders"), including those in rural areas. This has been done, e.g., in Kenya's Rural Access Road Program (RARP), in which local administrators, chiefs, politicians, and the like, have been involved in the road selection process, even though they do not fund any of the roads. Such involvement helps to ensure that any organization design serves the legitimate purposes of both those affected by the condition of the road network and those implementing maintenance activities. A specific time frame has to be agreed for achieving maintenance objec- tives (discussed below) but given the complexity of developing an effective road maintenance organization a long time period such as 10-15 years must be allowed for, as opposed to a normal project cycle of a few years. Alternative configurations of available and poten- tially available resources, and means of combining them, have to be considered along with an analysis of their likely costs, results and consequences. The alternative selected should be the most effective, not purely from an economic point of view but also in the sense that it will best capture the energy and commitment of those who will have to implement, and be affected by, road maintenance, that it will employ resources actually or potentially available, and that it will allow for those obstacles to road maintenance which cannot be removed, at least in the short run. It should also allow feedback since more people will be knowledgeable about what the maintenance organization is supposed to do. 3.10 Governments need to have defined as a long-term objective the size and composition of the road network to be maintained. As the MO usually maintains only the classified road network, minor rural roads of major importance to the communities served by them are likely to be excluded, and, consequently, there is constant pressure from local government and MPs for the MO to maintain some of the more important (to the community) unclassified roads. Government and the communities need to decide whether maintenance of some of the less important clas- sified roads can be dropped in favor of some unclassified roads. Long-term objectives should also focus on standards of road mainte- nance, for example, they should aim for optimal maintenance standards based on economic criteria,8 extend the scope of maintenance to include the gradual improvement of roads, such as installation of additional culverts, and the like (para. 2.04), taking into account 8/ As determined, for example, by the Bank's Highway Design and Main- tenance Model. likelv financial conscraints- Pclc guidelines need to clarify rhat should be done about roads that have deteriorated but that are still meant to be maintained. Artlher v the government should con- sider isý -.*e desirab_1ii,ty. of maaeac ciiisgenerating a- nany employ-n,r :pportunities is b and Ce stribu:ion of th'cse opportlnit4ez throughrCc the csuntry'. 3.11 Str-teey and Sjr1ture. The strategic p n process coers all technica1 and organ izati aspects cf t--e desi o a road i,n- tenance organization. organizational detres i -e not ra caeable of ,ervly affecting atnc tivities thereftre sb-tegies hue t be designed t3 avde : ir obsta to roac maiSeln-ce. Such actors include e `ical intet- ference ar- :, effect of te civil service ary.'r ativation. In sonm zase : will '% ,pn, ay for the e. es olf to be changed >t b- :dfied r m maintenance ert be effec- tive. What m;t be a-1e:i a inoring se e Ievelopme of maintenance capacitr, s is occurred se nternalLv financed maintenance pre:v:. A review of .s and af alternatives to deal w em follows. 3.1.' 2olitical Envir.: . When road ma aee entralized in iandia, the desigr . aintonance ortuat based on prc.-ams of -ork being a out -ecordine st naintenance engineering tritaria. mavce was made 'ýr t v politi- ized nature 3f t-e roact m, (s c.:urs ir :her 9E the wor<4 or for the inevi:able p tcal prsures an inter that affect roa,.ý maintenanc- etiti-s, As a re;ult, d bridges have been constr,cted .r u rade W.th maine:anc and mainte- nance -ehi les and equ rc r have -een used Lic works (and possi:ly private "equests" fro- YPs senior politicians, as well the Provincial dmin are chan- nelled hrough headqua-- staff and PEs, ahizle ntndents and ins-ectors are su' !lient pressu.re bo tIv from loc.-1 caiil nd ithrough t!.elr supez t' in1s- try. Te extent Co : z,e politiCa emands r t the wishes of the -ajority of th s living 1.n .oca, not known _nd andoubtedl; r o one a : o anot*,,.,- :oes the extent to which suc -a-n enance ac:--i"ies are ncry. Cer- tainly. near electices 'hern is strong political press-rie fýor roads sini. in order to be -aented, MPs struggle co jbtain more res as for their ci---ituern:is. - 13 - 3.13 Although some suggestions to curb excessive political pressures are proposed in the next section, undoubtedly there will always be political pressures that cannot be constrained or turned to the organization's advantage. (In fact, it could well be argued that the maintenance organization is the most suitable entity to carry out these public works.) Therefore, unless external agencies giving sup- port for maintenance activities can persuade the government to provide a separate fund, upon which the MO could draw for "extra-maintenance activities," realistic figures for extra-maintenance work should be allowed for in their estimates of recurrent maintenance funding requirements. It is important not to ignore such activities are tak- ing place and subsequently provide unrealistic funding estimates. 3.14 In the case of maintenance of feeder roads, programming work and performing maintenance activities require strong coordination links with the local environment, otherwise informal channels of com- munication will influence maintenance activities or the local popula- tion will feel frustrated and unsatisfied with the MO's efforts because their "felt needs" have not been taken into consideration. Political support for road maintenance needs to be gained at the national (such as the Ministry of Agriculture or Tourism), provincial, and local level (including administrative officials, chiefs, farmers, cooperatives). One of the main roles of the manager in the road main- tenance organization is to identify, and obtain support from, indivi- duals and agencies (such as local chiefs and cooperatives) interested in road maintenance and to involve them in the organization's activi- ties through coordinative mechanisms. 3.15 Centralization or Decentralization of Feeder Road Maintenace. Historically, under the local council system (para. 2.06)maintenance of roads was determined by negotiation between the roads staff and various local representatives (particularly the Works Committee of the Council), including the chiefs. The role of the road officer was to balance competing client demands. This approach has been supported by the "representative school" which stresses detailed knowledge of local needs (which elected representatives are assumed to have but mainte- nance officials are not). The problem with this argument for road maintenance is that some activities, such as routine maintenance (with the possible exception of grading), will not be deemed as necessary by the local community and its representatives (para. 3.06) and, there- fore, local maintenance organizations will tend to concentrate on re- graveling and improvement works. There is, however, always the possi- bility of educating them about the value of routine maintenance. There is also the problem (referred to in para. 3.04) of to what extent local authorities should follow clients' wishes (which may not be of high economic importance but of social and/or political conse- quence) as opposed to interests (determined by the professional's criteria and technical expertise).9 9/ P. M. Blau and W. R. Scott, "Formal Organizations" (1963). - 14 - 3.16 By contrast, putting the responsibility for feeder road mainte- nance in the hands of bureaucrats (local or central) not directly answerable to elected representatives may result in an organization insensitive to local road needs; thus roads that may be of high socio- economic priority to local communities might be ignored. This appears to be a problem in the case of Zandya, although local councils that still have revenues for road maintenance also appear to show the saIe insensitivity in some cases. Diversion of funds to other purposes, such as the creation of jobs with the local councils and diversion of equipment and funds to private use, is a common complaint. 3.17 As an alternative, one might examine the suitability of involv- ing other provincial organizations in road maintenance decisions. For instance, in addition to the Provincial Commissioner or Governor, where most of the power lies, and the chiefs at the local level, new opportunities might be offered at district levels, especially whi.re government has organized special development councils composed, ior instance, of representatives of the different government service agencies, MPs for the district, district administrators and community representatives. 3.18 Stronger links might be established between the central goverr- ment-controlled maintenance organization and the local developmen councils, which could be made representative by increasing the nunber of elected members or by eliciting recommendations from separate road-user committees which could be formed at the sublocation level, comprising elected representatives, together with chiefs, agriculture cooperative and extension staff, or other local officials. If the idea of elected representatives is not acceptable, then local of-fi- cials together with local appointees could form the road-user coivT- tee. There is also the possibility of making use of self-help loc:al committees. The local road user committee could then be involved Vith the MO in programming maintenance work, with the MO reserving a cm- tain amount of limited resources for priority routine and pricrity periodic maintenance activities. As part of the educational proc;s3, the MO should explain to the committee the necessity for these activi- ties and only allow the committee to select other priorities. If 'he committee decides that certain unclassified roads should be main7ai1ed because of their local importance, then the MO could accept thei- decision (with some possible exceptions) as long as the committee v,re able to say which classified roads not to maintain. This would instill a sense of cost-consciousness and encourage the local population to think in terms of relative priorities and trade-of.-. Local representatives could be informed when funds, transport, 0- equipment were not available and could put more pressure on gove-7-ent to obtain the resources. Alternatively, they could try to raise them locally. These resources would include, for example, funds for a spare part for a grader or truck, funds for hiring casual labor, or contributing labor itself, or the loan of transport from the coopera- tive or other source. The possibility of raising local resources will - 15 - depend very much on the history of self-help projects in the area, and the interest of the local leadership. In some districts of Zandya, for instance, bridges and small sections of roads are being built through self-help projects mostly with money raised to pay casual labor. 3.19 Under the system in which a local committee is involved in the programming of work, when Headquarters or provincial staff receive an MP's or other prominent person's request to gravel certain roads, the request should first be approved by the committee. In this way, requests that serve only individual interests can be avoided. When agreed programs are curtailed and funds are shifted from one district to another, some explanation should be given to the local committees. Local awareness of road maintenance requirements may also create more local pressure on the road maintenance organization (in particular, to improve the MD's performance), as well as the provision of a program- ming and feedback system. The inclusion of small rehabilitation and improvement works is a carrot to create local interest in participat- ing, and again, is something the MO is already doing in Zandya and will not cease from doing despite protestations from "the maintenance work only" school (usually foreign aid agencies). The proposals made here are an attempt to develop a strategy to deal with the political environment, which is already affecting the MO's activities, disrupt- ing work programs, and diverting scarce resources. They are intended to show those designing organizations possible ways of taking into consideration the obstacles in the environment and incorporating in the design of the organization those who are interested in road mainte- nance. 3.20 In order for the system of more local involvement to work, the PE must allocate funds to the district for the maintenance of low- standard roads. This allocation should be based on the PE's knowledge of local conditions. In practice, Headquarters would allocate funds to the PEs who would then allocate them to the districts, thereby elimi- nating the role of Headquarters staff in decisions about rehabilita- tion, improvement or graveling programs for the low standard roads, or in the implementation of such programs. With respect to the low- standard earth and gravel roads, the Headquarters needs to take a staff policy, as opposed to line, role, providing professional guidance to PEs on technical, organizational, and other matters, as well as playing an advocacy role for feeder road maintenance at the national level. All requests concerning feeder road maintenance would then be channeled to the appropriate level, that is, the district and sublocation level. 3.21 If these measures are considered not immediately applicable or turn out to be unacceptable to the government, how else can the main- tenance organization use the environment to at least gain more resources and carry out more work? One alternative is to seek the assistance of local development councils, chiefs, cooperative offi- cials, MPs, and other important local dignitaries in raising funds when there are none for hiring equipment or purchasing fuel and other mate- rials such as gravel. In exchange, the MO can offer the use of the - 16 - otherwise idle labor force and equipment on mutually agreeed road main- tenance activities. Where transport is not available, the MO could request local officials, such as cooperatives, for assistance -- again, in exchange for work important to those concerned. As one of the PEs interviewed is already doing, help can be solicited from chiefs to stop practices detrimental to the roads, such as blocking mitre drains, which prevents water from running off the road. If water from blocked drains is causing damage to the farmer's crops, their complaints can be channeled back through the chief. Similar strategies can be tried to make the environment work for the organization, but they require imagi- native, empathetic, and enterprising engineers, superintendents, or inspectors to initiate them, and the lack of such people is one of the biggest weaknesses of any organization. It may be possible, however, by constant retraining and seminar efforts, to create greater awareness of different strategies to follow (para. 3.38). 3.22 The Financial Environment. One of the most commonly cited rea- sons for the inadequacy of maintenance organizations is the lack of budgetary resources. For instance, in Zandya, budget allocations for road and mechanical maintenance declined or remained stationary in real terms in most years from 1974 to 1979, despite an increased workload, resulting in operations being halted for 3 to 4 months a year. Every time the economy goes into a serious decline, often precipitated by oil price increses, not only are road maintenance funds cut or not increased in real terms, but other financial restrictions are imposed that have repercussions on the maintenance organization. The most com- mon restriction that affects road maintenance is the imposition of import controls, which particularly affect the supply of spare parts in the country. Because this particular problem may be temporarily eased through financing of an inventory of spares by foreign aid agencies, it is less central to this paper than the internal measures that govern- ments may consider. 3.23 Recognizing that the Ministry of Finance's commitment to mainte- nance is likely to fluctuate, depending on the availability of recur- rent financing and on the overall government policy toward increasing the total recurrent budget at the expense of the development budget, various measures can be suggested to government to increase maintenance funding.10 But, whether government agrees to increase the recurrent maintenance budget by so much a year, or whether a separate semiauto- nomous Maintenance Authority is set up, which receives a set percentage of road-user revenues (as in Haiti), will depend on government's con- tinuing commitment to maintenance since at any time they can rescind original agreements. 3.24 An example of how financing arrangements can break down is the hire-charge system (para. 3.04). The introduction of a hire-charge 10/ For a detailed description of these measures, see World Bank "The Road Maintenance Problem", op. cit. - 17 - system for equipment in Zandya was meant to encourage the MO to make more efficient use of its fleet and thus reduce its overall equipment requirements. However, in the past it has resulted in a waste of resources because, although available, equipment has stood idle because the MO could not afford the hire-charge or the fuel to run the equip- ment. If the system had been more flexible, the hire rates would have temporarily excluded the depreciation component, while still including depreciation in a separate cost accounting system, and, with the lower rate, MO could have purchased more fuel and utilized the equipment to a greater extent. But this does not avoid the problem of funding for replacing equipment. 3.25 Other ways in which the maintenance organization could attempt to raise more funds for road maintenance are: (a) financing mainte- nance activities, such as reshaping of roads and restoring drainage ditches through construction and road improvement contracts, that is, every time a contract for a gravel road is let, a certain amount of other minor maintenance work on surrounding priority roads could also be undertaken by the contractor, and if the contract is donor-financed it may be possible to persuade the donor to finance the additional maintenance works; (b) including in special development programs, such as the construction of feeder roads, routine maintenance activities under the organization which is going to be ultimately responsible for maintenance (discussed in para. 3.50); (c) requesting donor financing for maintenance equipment and spare parts for the equipment's life expectancy (as under certain loans by the Canadian Aid Agency, CIDA), or donor financing for maintenance of equipment by private dealers (to relieve the MD of some of its workload) and financing spare parts as a project component of road construction projects (on the grounds of improving the maintenance capacity to maintain adequately the newly constructed road and adjoining roads). The rationale behind such an approach for donors is not the purpose of this paper, but one of the supporting arguments is that donors are financing maintenance anyway, through the far more costly form of rehabilitation projects. Also, the argument that donors should finance only development activities with a longer period of economic return will not be valid in the case of roads constructed under low-class programs, which are likely to deteriorate in a few years instead of lasting the projected 10 to 15 years. 3.26 These alternatives are only a few for the MT and donors to con- sider. It might well be that none are feasible, but they are primarily meant to illustrate the flexible approach any maintenance organization needs to take in order to obtain more funds. Other possibilities of raising money locally, should, of course, be considered, but in many cases there will be little prospect of raising money for maintenance of feeder roads through local organizations such as the local councils, since very little revenue is raised for any activity at the local level. The only possibility would be the introduction of a special cess (tax) on agricultural produce marketed, but that would be an extremely unpopular government decision. Voluntary funding is likely only in certain instances rather than on a continuous basis, since maintenance is a recurrent activity. Experience so far in Zandya has - 18 - shown that money can be raised for capital projects through self-help movements, but there is little willingness to raise money for continual operation or maintenance of facilities once constructed (except in swampy areas where villages performing the maintenance works are allowed to levy a toll from road users). That puts the burden back on central government, which means road maintenance must fight for scarce resources along with every other sector. If funding remains inadequate despite all attempts to obtain more, then naturally the maintenance program has to be reduced. The extent of the reduction plus the eco- nomic implications need to be made clear to all concerned. At present, however, even more important than the financing problem is the use of existing funds, for if they were used more efficiently road maintenance could be substantially improved. 3.27 Improving the "Controlled" Environment: Motivation and Incen- tives. As mentioned in para. 3.01, considering what resources the maintenance organization has, productivity is still low (especially for such routine maintenance activities as grading). One reason for the maintenance organization's poor productivity is that it operates within the civil service structure (para. 3.05). The problem is compounded by the fact that civil servants are allowed to own private businesses, except when this is in direct conflict with their work; therefore many government employees run small businesses. High-level civil servants generally own large farms and some are leading businessmen in their society. Undoubtedly, government-paid time is sometimes used for private purposes. 3.28 Some alternatives to consider are: (a) to try to reduce the amount of work carried out by force account by contracting out more work (paras. 3.39-3.41) to avoid the problem of working with unmoti- vated civil servants; (b) to do force account work but with casual labor instead of permanent government employees (paras. 3.42-3.44); and (c) to alter civil service regulations to introduce more market incen- tives or to bend the current rules as much as possible. The following is a discussion of some of the alternatives for consideration with respect to (c). 3.29 With a permanent work force, governments will need to find ways of motivating their personnel to increase efficiency. Unfortunately, little work has been done on this subject in the African context11 and western management concepts may not be applicable. Considerable research assistance could be given by external donors and local univer- sities by carrying out studies in various African countries to deter- mine the conditions under which greater productivity is realized. 11/ Notable exceptions are D. L. Leonard's "Reaching the Peasant Farmer" and C. C. Onyemelekwu's "Men and Management,". See also, T. A. Lambor, "Motivational Factors in Work" (paper at 7th Conference); and Samuel Paul, "Strategic Management of Public Programs," Harvard University and Indian Institute of Management (forthcoming). - 19 - Short and simple surveys could be carried out in private industries, especially with successful road contractors. Even within Zandya's Roads Department, there is ample scope for surveying different road development programs (or parts of programs) that are successful and un- successful to determine the relevant factors leading to good perform- ance. One area to study is the extent to which money is an incentive of workers when tied to production. Many private contractors provide bonuses to their workers for good work and it may be necessary for governments to consider seriously the introduction of bonus systems as part of the civil service pay structure, especially where large invest- ments are concerned. Other factors to look at are the ratio of super- visors to laborers, how tasks are split up, how teams of workers are put together, and so on. 3.30 External donors have tended to concentrate on the project approach, which often limits them to assisting one sector, even one institution at a time. Therefore, the problems common to many institu- tions are not clearly recognized, or are identified but cannot receive the necessary assistance within the context of a project. Such areas as government remuneration and regulations require considerable assistance, especially budgeting and procurement procedures. Incentive systems incorporating bonus systems would necessarily have to be tied to easily measurable activities, such as kilometers graveled and number of cul- verts cleared. Better use could also be made of the promotion system (again requiring changes in government regulations), although in a context involving ethnic, linguistic, seniority, or other personal or social considerations a promotion system based essentially on merit may be more difficult to administer. Other incentive schemes require few changes in civil service regulations and are already used by a few managers in road maintenance. These are overtime or time off for per- forming specified tasks on schedule (as observed in Sierra Leone), recognition of good work in departmental newssheets, and competitions with prizes for the best work performed (for example, team and indi- vidual prizes for "mechanics of the year," with awards made at the annual national agricultural fair). Such incentive systems make it possible to put more group pressure on individuals to perform - espe- cially on mechanics to keep equipment running. 3.31 The research carried out so far in Africa and other parts of the world indicates scope for the sociotechnical approach to organization design, since workers appear to receive greater satisfaction from their jobs if social interaction is possible and if they are given a "whole" task to perform. Again, studies or experimentation are necessary to see to what extent the team approach can be applied to maintenance activities in a particular country. One possibility is to provide equipment units with their own mechanics, as fully equipped as possi- ble, and to give specific tasks to perform and incentives for meeting targets. This approach would mean a certain level of duplication and underutilization of mechanical equipment but the costs of duplication would have to be balanced against the benefits of having independent units (just how independent would have to be determined by trial - 20 - projects). Teamwork will be possible only if mechanics and supervisors assigned to teams work on specified items of equipment thereby creating a sense of identification with their equipment and with mechanical teams looking after equipment in the field. Zandya's present system isolates different functions, so that different groups such as the road camps and grading and regraveling teams have no control over the mechanical units responsible for maintenance of their equipment and vice versa. Also, considerable programming skills are required to bring equipment from different units into workshops to be repaired and returned to the units. 3.32 The same argument could apply to the provision of workshops. At present, Zandya's maintenance organization depends on one central work- shop for overhauls, one regional workshop for other major repairs, and district or local workshops with only basic tools for simple repairs. Closer attention could be given to the type of facilities successful contractors have for the maintenance of their fleet, since they face the same trade-off of comparing the cost of underutilized workshop equipment and other resources with the cost of waiting for repairs to be carried out at a distant facility. Since government organizations are not as efficient as private contractors, it might well be less costly for the maintenance agency to have better equipped district or local workshops. 3.33 Another incentive for better performance is the assigning of specific tasks to employees. Once an employee has mastered his spe- cific task, it is important that he not be transferred (as is too often the case) since this only results in loss of morale and decreased pro- ductivity. Workers should, of course, be encouraged to seek more advanced training so that any transfer will be to a better position. If the job has a technical content or requires special training, so much the better, because it raises the social status of the operator -- and the incentive to perform well in order to retain that status. In the Zandyan context, a worker made to take work of lower status than his normal job (even without officially demoting him) loses face. This should, therefore, only be done as a form of punishment for poor per- formance (probably one of the most effective punishments and one which is in limited use already, for example, assigning drivers to work as flagmen because of poor performance). 3.34 Supervisory Staff. Without adequate staff to provide guidance and motivation, no incentive system will be viable. Supervisors work- ing directly with laborers, especially at the overseer level, fre- quently comment that they learned techniques at training school but not how to manage people, perceived as the most difficult part of their work. This is because training takes place in ideal conditions in the training school and in the training brigade, with supportive col- leagues, all the necessary resources, and close supervision by instruc- tors. The enthusiasm of candidates entering training school is often remarked on, but somehow this spirit is dissipated once the trainee goes to the field, where he tends to be rather isolated from other parts of the agency, receives limited support and necessary inputs, and is often faced with managing too many maintenance workers. - 21 - 3.35 One of the problems may be that supervisors are taught too much in one training session, and that together with the complexities of the real world as opposed to the training school, expectations are high about the work they can perform. Supervisors may first need to prove their competence at a very low, clearly defined level, such as super- vising labor doing only culvert, ditch, and bush clearing before pro- gressin to more complex activities such as grading and graveling works This would be in line with task specialization discussed earlier (paras. 3.31-3.33). The same principle of progressive training could apply to mechanics and inspectors. Instead of covering all aspects of mechanical work, mechanics should learn only about one or two makes of vehicle at first and gain experience with those before progressing to other vehicles and equipment. Trying to stretch newly- trained mechanics over the whole fleet only results in low produc- tivity. 3.36 The TD in Zandya has been an excellent training institution and has produced large numbers of trained personnel. But it is also con- cerned about the performance of trainees in the field. Consideration should therefore be given to TD providing more support in the field after training programs are completed. By getting involved in the field, TD can adapt training programs to the problems supervisors meet in everyday work and advise on recurring managerial problems. TD sup- port in the actual work environment is better than the training bri- gades', since many of the newly trained supervisors do feel isolated (particularly in the larger provinces). The addition of a cadre of "advisors"from the training school, responsible for helping the over- seer or inspector in the field program and in supervising work, would provide the necessary follow-up and close supervision. 3.37 Whether more supervisors are required or not will depend on their productivity, but if there is a need, it should not be an insur- mountable problem to train more maintenance supervisors (as well as instructors), since Zandya has such a large number of unemployed sec- ondary school graduates as well as educationally less qualified per- sonnel who have good leadership potential and maintenance experience, such as drivers, equipment operators, and artisans. 3.38 Whenever management training is discussed, very little attention is given to maintenance and mechanical engineers. Yet, as pointed out in para. 3.34, maintenance is more a managerial activity than an engi- neering one. Nevertheless, engineers with no management experience are assigned to maintenance and technical departments, and because of the pressure of the workload, there is little time for anyone to train them or even notice if they are performing poorly or well. This suggests that training schools, or institutions such as management schools, 12/ More detailed suggestions can be found in P. C. Ryall, "Training for Road Maintenance in Developing Countries", ORT, Ninth IRF World Meeting, Stockholm, June 1-5, 1981. - 22 - should design special courses not only to instruct engineers in pro- gramming, budgeting, and procedural systems, but also to instruct them in motivational management techniques: how to understand the environ- ment in which they must operate, how to attempt to influence that environment, and how to set up formal or informal coordinating mecha- nisms. The MT could pool its resources to provide a management school. Many of its activities require considerable managerial skills, and involve substantial maintenance activities requiring the same pro- gramming and people management skills. 3.39 A continuing problem for the public sector in Zandya has been the lack of supervisors and engineers, together with a shortage of mechanics who are attracted to the private sector by higher wages. Through the efforts of the TD, however, these problems are gradually diminishing, but the problem of the quality and motivation of the staff still remains. Since the staffing of the organization with a large cadre of experienced people will take time, this has to be allowed for in the design of the organization, that is, the organization should be designed around realistic estimates of staff (and other) constraints, and in estimating the capacity of supervisors to achieve maintenance objectives. Choosing Alternative Structures 3.40 (a) Contract Maintenance. Since scarce resources must be maximized in countries like Zandya, it is necessary to develop maintenace capacity, not only with- in the maintenance organization, but also in the private sector. By developing local contractors, the maintenance organization relieves supervisory staff of activities such as procuring and delivery sup- plies, organizing staff and equipment in the field, and motivating staff; in addition, it relieves the organization of some of the burden of procuring and maintaining equipment. Of course, the organization will still need to supervise contracts to ensure that works are car- ried out according to contract. 3.41 In Zandya, the relative lack of sophistication of local contrac- tors makes it possible for the maintenance organization to exert a strong influence over them (as part of the training experience) and to control prices (normaly unit prices are fixed by the maintenance organization in agreement with contractors who do not have the experi- ence to cost out and submit bids). Local contractors may not perform as efficiently as foreign contrctors, but more important, in comparison with most government maintenance organizations in Zandya they have per- formed well. With the increase in funds in the last two years, re- graveling carried out by local contractors has been the only mainte- nance activity which has substantially improved. There is further potential for developing more local contractors not only to gravel and rehabilitate roads but also to carry out routine maintenance activi- ties, especially since there are many entrepreneurs in the country and - 23 - an environment which does not discourage private enterprise. Any pro- gram to develop local contractors will meet with a high failure rate (although no more than by force account) and will take many years; therefore, this needs to be built into project estimates. Also, at least in the first few years, some guarantee of work must be given to new contractors, since in the past many efforts to encourage them have failed because the government could not provide them with sufficient work, and because they had not yet learned to diversify, they went bankrupt. Even in the United States, the trend is toward disbanding large maintenance organizations in favor of contract maintenance. 3.42 (b) Casual labor With respect to its own organization, the MO must decide whether to retain its fixed work force or contract casual labor as and when necessary. The problem with a fixed government force is one of motivation (paras. 3.29- 3.33), whereas casual labor may be more finan- cially motivated and can be dismissed if their work is poor. The greatest advantage to MO using more casual labor and keeping only a small, highly trained permanent labor force for activities requiring skilled labor (grading, bridge repairs, and so on) is that costs can be reduced in time of budget constraints, since at present the MO's per- manent employees cannot be laid off. Casual labor can also be obtained at less coast than fixed labor if road camp, uniform, insurance, and other costs are taken into consideration. The real cost of permanent labor is even higher if one takes into account the amount of time the labor is idle because of lack of transport and equipment. 3.43 The disadvantage of casual labor is that each different group of laborers must be trained in maintenance techniques and their activi- ties supervised. To what extent more supervision is required for casual labor as opposed to permanent labor is not clear, since, in Zandya, experiments have been carried out only with maintenance over- seers not used to supervising casual labor. More research is needed on the supervision problems of casual labor employed in maintenance; much can be learned about the training of supervisors from the ongoing labor intensive feeder roads programs (LIFP) in Zandya. By carrying out regular annual programs of such simple activities as cleaning culverts, drains, and drainage ditches, and bush-cutting, repeated training can be reduced as the local population gains experience. Initially, while the MO slowly runs down its permanent work force through natural attri- tion, groups of casual laborers could work alongside, and thereby gain experience from, permanent employees, which could result in greater productivity. The success of employing more casual labor will deter- mine the extent to which the permanent work force can be run down. Experimentation should be carried out and results carefully interpreted to determine whether to replicate this approach. - 24 - 3.44 The other alternative is the length-man type of system whereby casual labor carries out routine maintenance work, with each worker contracted to maintain a few kilometers of road along which he resides. The success of this system will vary from one area to another, but different systems can be tried to suit local conditions. In some parts of the country, the best worker could be paid extra to supervise the work of others. The next stage is to transfer the system to the maintenance organization, but if some of the factors previously mentioned do not improve, the system could collapse through inadequate inspection and diversion of funds. The lengthman system has been used on rural feeder roads with fewer than 20 vehicles a day on well- constructed roads built by local labor. In Zandya, one attempt to replicate the system on existing classified roads with higher traffic has failed, because workers could not be induced to adequately maintain their section of the road. Unfortunately, the scheme was not closely monitored, and in this situation much depended on the quality of super- vision, especially since the labor contracted to maintain the road had no previous construction experience. Nevertheless, the length-man sys- tem should be tried in various areas of the country with well-designed pilot projects. If found successful, it could be implemented by the MO on classified low-class and low-trafficked roads in areas where labor is available at the casual wage rate and as the permanent labor force is decreased. 3.45 Choosing an Alternative Technology with Different Organizational Implications: Capital vs. Labor-Intensive Maintenance The major problem of applying capital-intensive methods to road maintenance is the low utilization and availability, and abuse, of equipment (para. 2.13). Typically, depreciation rates in Africa are several fold those in Europe and the United States. Depreciation is not, however, the major cost; low availability rates severely affect the productivity of maintenance units, resulting in increased costs per km, or a complete standstill of equipment units, which leads to further road deterioration. Essentially, the problems of maintaining a fleet of equipment are organizational and managerial, as well as financial, requiring more trained and experienced mechanics and supervisors, availability of spare parts, and replacement of aging equipment. If insufficient funds (including foreign exchange) are provided for spare parts, or import restrictions limit the supply of vital spares, a mechanical organization cannot perform its task no matter how good the management. 3.46 Possible alternatives for Zandya to overcome some of its prob- lems with capital-intensive methods are (a) to request donors to finance spares over the life of the equipment they finance; (b) to pro- cure equipment over several years rather than at one time, so that the whole fleet does not age at the same time; (c) to take on more tech- nical assistance for the MD; and (d) to contract out maintenance of part of the fleet, which in many cases is no more expensive in economic terms than the cost of an underutilized equipment fleet. - 25 - 3.47 A further alternative to be seriously considered is keeping the size of the equipment fleet at the level at which it can be maintained by MD or more significantly at present, at the level at which MO can adequately utilize the available equipment. This implies the use of more labor-intensive techniques. The LIFP has considerable experience of labor-intensive programs in Zandya, from which many lessons could be drawn for road maintenance. As well as routine maintenance, other activities such as regraveling could be done by a mix of simple equip- ment (tractor-trailer) and labor-intensive methods where the haul dis- tances are short. Longer-haul distances could be contracted out to private tipper operators. The arguments against labor-intensive methods are low productivity, requiring intense supervision, and lower quality of work. The latter need not necessarily be the case, espe- cially if simple equipment is provided (for example, rollers for com- pacting). Supervision for labor-intensive techniques differs from that required for capital-intensive techniques only in that the former requires more supervision on site whereas the latter requires more supervision in mechanical workshops. 3.48 In times of fuel, financial, or foreign exchange shortages, labor-intensive maintenance could proceed while capital-intensive main- tenance would come to a standstill (as occurred in Malawi during the 1979 fuel crisis). In addition, the cost of poor-quality supervision is less for labor-intensive maintenance than for capital-intensive maintenance. In the former, only the cost of underutilized labor is incurred while in the latter, high costs are incurred by broken-down equipment (because of operator abuse or poor maintenance) and by idle equipment in need of repair and fuel. Labor-intensive maintenance also increases employment and decreases consumption of fuel. 3.49 The Integration of Development and Maintenance. One of the problems the maintenance organization faces is limited coordination between its activities and those of the Construction and Special Proj- ects Branches; it is not uncommon for the MO to be unaware of proposed design standards for certain roads until after they have been decided and therefore have no say in the decision making process. Road con- struction programs are implemented in isolation from the maintenance organization. Foreign aid financed projects may therefore be putting more work on the maintenance organization than it can handle. 3.50 Although the Bank and other external agencies often consider that money has been poured into road maintenance with little lasting result, the same could be said of some aspects of road development pro- grams. Consultant services and other technical assistance, plus equip- ment, are provided for building roads, yet once the roads are com- pleted, little of the experience and organizational ability is left behind, and often the roads fail prematurely because of poor mainte- nance. Millions of dollars are spent on consultants to design and supervise road construction, yet there is often not the same readiness to fund such assistance for building up the long-term capacity of an - 26 - organization or the local contracting industry to maintain a multimil- lion-dollar road network. More could be obtained from road construction programs by incorporating road maintenance components into them. For instance, when feeder road programs are undertaken a more lasting effect could result if, at the same time, a maintenance capability for the roads constructed and for the existing classified network connecting to them were developed within the organization ultimately responsible for their maintenance. Even if the extra cost of the maintenance components meant building fewer roads, it could be economically worthwhile since maintenance has such high returns compared with development. Too often road improvement programs completely bypass existing government entities such as the maintenance organization; subsequently, once the roads are completed, there is little in the way of institutional development for maintaining them. 3.51 Administrative Environment Zandya suffers from the usual "red tape" that characterizes many developing countries' bureaucracies, inhibits efficiency, and occupies qualified personnel's time. The procurement and tendering procedures have held up construction of buildings, the acquisition of vehicles and equipment by the MD and, most seriously, the repair of vehicles and equipment. Strict tendering limitations have meant that the acquisition of spares costing little money has had to be approved by the Ministerial Tender Board, and some major spares costing thousands of dollars have had to be approved by the Central Tender Board, which takes weeks. Con- tracts awarded for the supply of particular spares have eliminated the need to continually refer to a Tender Board for approval, but prob- lems still occur when spares are not available from the contracted firms. According to government regulations, the purchasing officer must obtain documentary evidence that such spares are not available, but this is not easily obtained since no supplier wants to admit failure in adhering to the contract. The MD is gradually overcoming these prob- lems, as well as the problem of a general shortage of spare parts in the country, by holding their own stocks of spares obtained in bulk -- but at considerable cost. Thus, they have attempted to bring within their control the problem of spare parts, but it remains to be seen just how difficult they find the management of a complex supplies system. 3.52 The administrative framework of the government in Zandya requires considerable overhauling to eliminate the need for agencies such as the maintenance organization to be hampered by excessive paper-pushing and by lengthy procurement and other processes. The trade-off between the introduction of regulations to prevent corruption or misappropriation, and the need to let government agencies operate efficiently is hardly ever made. Indeed, those who draw up the regulations are often far removed from such operational agencies as the maintenance organization and either have little knowledge of, or do not consider important, the effect regulations can have on the efficiency of operational organiza- tions. - 27 - IV. CONCLUSION 4.01 The foregoing recommendations are alternatives that can be tried out by many African countries with assistance from external donors. They are predicated on the fact that experience has shown that tradi- tional organizational approaches have not always worked in practice over the long run. The recommendations are not proposed as "solutions" but as a basis for pilot projects or studies to determine if they are appro- priate for the set of conditions described in this paper. Different combinations can be tried out, but the effectiveness of any will depend on the resources available in the country and those provided by external agencies. Above all, the environment in which road maintenance is to be carried out has to be appreciated before the design-stage of organiza- tional development can take place. If the environmental constraints put on the development of road maintenance capacity are not understood and allowed for in organization design, then the subsequent effectiveness of the organization will be severely hampered. Hence, it is our hope that at least the method we have chosen to analyze the problems of road main- tenance organization and management can be applied in many Sub-Saharan African countries. Chart 1: ORGANIZATION OF THE MINISTRY OF TRANSPORT (MT) WITH RESPECT TO ROADS Permanent Secretary Engineer-in-Chief Provincial Roads Department Mechanical Training Engineers (PE) Chief Roads Engineer Department (MD) Department (TD) Maintenance Construction Organization (M0) Branch Special Projects Chart 2: STRUCTURE OF THE MAINTENANCE AND MECHANICAL ORGANIZATIONS Chief Maintenance reCi Mecnical Engineer (MO) Engine- 2 E) ^:.ineer ND Provincial P ro cial Road3 Mechanical Engineer (PRE) EngI'-leer (?ME) (Prcvinc4al Works-op) Divisonal Roads Engineer (Region) Suäerintendnt Superintendelt (Regior) (District) 9:,1ior Ins :r Ser r ' cr (Distr--t) (Division)' nspec . Road Camp Foreman or Annex Page 1 ANALYTICAL FRAMEWORK FOR ORGANIZATION AND MANAGEMENT OF DEVELOPMENT PROJECTS1 1. Introduction 1.1 Design of development projects based on conventional organization theory results in organization structures and processes that often prove inadequate during the project implementation stage; and many of the dif- ficulties currently labelled management problems, are in fact problems of organization design. A complementary/alternative conceptual orienta- tion, which simultaneously addresses both internal and external organi- zation and management issues, is needed. It is not enough to relate a project organization to its environment; instead, the organization must be capable to interactively (and adaptively) deal with other organiza- tions and the environment itself. In development organizations, project success is determined not so much by the most logical or efficient arrangement of internal organization and resources, but by an appropri- ate co-alignment with external agencies (local, provincial and national government agencies; input suppliers; output users; and -- most of all -- project beneficiaries). 2. Understanding the project environment and organizing it to benefit the project. 2.1 Before designing the organization, it is necessary to identify and understand the environment in which the project will operate. The environment is not "everything out there"; there is a level between the organization and the "uncontrolled" environment which is external to the organization but can be influenced by the project's management. There- fore, project designers and managers should be concerned with three levels of environment: (a) internal or "controlled" environment enclos- ing the elements directly needed for productive activities, and defining the organization's boundaries; (b) transactional or "influenceable" environment consisting of external entities whose activities can influ- ence organizational and management performance; and which in turn can, and should, be under the mutual influence of the focal project organiza- tion; and (c) the contextual or "appreciated" environment, comprised of institutions (social, economic, political, technological and cultural), that produce activities affecting organizational performance; and that can neither be controlled nor influenced by its management, but must be understood and valued instead. 1/ For details see World Bank Staff Working Paper No. 375, op. cit. This summary note was prepared at the request of F. Hotes, Irriga- tion Adviser, World Bank, by Pammi Sachdeva and William E. Smith, The Wharton School, University of Pennsylvania, Summer 1980. Annex Page 2 2.2 It is never possible to completely specify the elements of the appreciated and influenceable environments in advance of pcojecr or organization design; the very act of design and implemenracior. Uncovers new sources of influence and adds to knowledge of these environments. Also, for any project, the relative importance of each en ironmnat dif- fers. In single sector, physical-infrastructure projezts fuil1ding a dam or a major highway), the relative importance of the eerolr:led environment would be high; for projects with a broaefr orientation (man- aging large irrigation systems concerned with equiral is.0iuttion and productive use of water, maintaining a large network' 'ander roads), the activities under control are relatively fewer. ir n as a project evolves through design, construction and ope i i* phases, it is achieving certain results. These very results obti ,1 *1cer the pre-existing patterns of control, influence, and a ct. Hence, there is a need for an organization design that is s ary flexible for adaptation to environments that are likely to ce. 2.3 In sum, the first dimension of our conceptua. , is a more precise definition of the organization in relation :a e.vironment. The key insight is the addition of the "influenceable e resent". The implications for projects' management are profound. Th2 7ir.agement role can no longer be seen as primarily inward looking The %nazer must focus not only on those internal elements subject to- h!s s trol, but equally -- and often even more -- on those externa l: of the environment that are subject to his influence and ti-at `t ai tc appre- ciate. 3. Building an organization based on participants' comm. set towards its purposes. 3.1 The second link in our conceptual framewor: is A e on the fact that this relationship between the organization an. its =wdronment has a purpose/political (or power) base. Organization-- are strumaces for bringing together people who see their involvemeni: as a iaars of a.chiev- ing their purpose. Participation is always partial (the organizafi!on satisfies only part of the individuals' total purpose), :td condi_i'Xal (on the organization being able to continue to su tr.y te inducamones necessary to the participants' continued contrt!-ion). Therefor., it is impossible to design a project unless we clairly map ou: its, a: its potential members'/beneficiaries' purpose. The Y steps in this -:san- tial political process of organizing are: (a) the establishment of a purpose, whiz: gives direction for the exercise of power* (b) the development of an appreciation, e, an understanding and valuing, of the o-o-rtuni- ties and risks in moving in the seleaed direction; Annex Page 3 (c) the development of a network of influence, i.e., joining with others whose own purposes suggest the possibility of mutual transactions that will aid in the achievement of the selec- ted purpose; and (d) the choice of actions and resources that can be controlled to achieve the purpose. 3.2 Each of these steps represents an increase in the degree of power (from appreciation, to influence, to control). The art of designing development projects is that of devising a network of purpose and influ- ence among organizations and individuals that will ultimately lead to a set of controlled activities producing the desired -outputs. The power process is important at all levels -- national, provincial, organiza- tional and individual; and must be recognized as such. Project designers should explicitly take into account the fact that the career- seeking motivation of bureaucrats' behaviour is designed to increase their own visibility rather than project performance. Similarly, polit- ical maneuverings are based on personal desires, motivations, and per- ceptions and should not be excluded from the formal design effort because they are somehow "irrational". 3.3 The practical implication of this viewpoint is that purpose, power and politics are recognized as crucial design variables, rather than as "undue interference". The process of development, defined as "the beneficiary increasing his control and influence over, and his appreciation of, his environment", itself has a political dimension. The project designer and manager can succeed only if he is able to appreciate the "political" field as it relates to the project, and can actively harness it for the purposes of the project, its various stake- holder groups, and individual beneficiaries. 3.4 The problem of structural design, is therefore, one of linking the purposes of the power centers in a way that will ensure progress towards their consensus of objectives. Mechanisms have to be developed for controlling (where feasible) and coordinating both the internal and external component parts. Control can be generally exercised along the organization's vertical dimension, and here means that one level/agency can cause another to do what it wants. However, though there are degrees of control, some activities, especially those between agencies (along the lateral/horizontal dimension), cannot be controlled. These activities must then be influenced through various coordination mecha- nisms, ranging from ad hoc meetings (a weak mechanism), or committee membership (of moderate strength), to participating in matrix struc- tures with dual reporting relationships (a strong mechanism). 3.5 While the mix of control and coordination mechanisms necessary and/or feasible varies from project to project, some general guidelines can tentatively, be formulated for development projects: Annex Page 4 (a) at least -oerate power is required at all levels Cl-oe developmýentoraiton (b) relatvr srong atterns ii cnrol ad co- ordf:stin ar-- r-aîäred at a L; (c) a b-i-ncet relattahip is n etween the stre'::h of coclation and contr at the samic lvel; and (d) the most prois:n area for org:zaional design impr1s Is the use T .ore and improved cooritz'ntion mechanL:. 4. Desi ampopropri a. .e'Iarning meet g. vns: time as a design dimeng ey 4.1 It h: 3d 4t nott _, rver, that h icial organizational requiremenr s nce just e rght" mix o ectanisms and processes, but the ability to moify i. tonment(s), .A to aiapt and change as necessary. fhe requeired ,, ngcum-eval in system for self- diagnosis and corre-cicn e , on the -,-w jJ frmation between levels. Th flow Ls g 13hy the detm , contr,l a-d coordi- nation sy w wich it "nditiord "tricution of power. Ec e;rtl of pv f eavelopmen- ae insance, or a highway or .t-Lation haý 'inct cont l: lake Ln this pro- cess, and a_, - distinct ,on for i -, (nacional - legitimacy acermediat :tiveness; cincy). Project evaluatio ,herefore, that the cga: trace its decisions ,-ards thrv, levels of structure to its purpose a-i it reatn tio its environ-a-, 1: cipated cos- que,ces i- . :t e proble-' sinL r or stra t'r due to va71bhs in the "i: :iceable" n ,ciated" env,:: 4.2 This in>lies chat vzati_onal leara:I _-uld 'e a coii- nuous pr( : It depenc _factive manage f aforIaticr withyin I~ -anizaýton t ex'.char,ge with i- v a rnm-i t, and 1.s thu-, at t:gal part 3 rganiza:iac desi : such a lami system, level exercise- ric hierarchical ei :1 over anv c" -- each -:-vel is closer tz influencing the othe: - vidä the process is more cne of coordination en semi-autonomous l It is < it- ical process of confrcntari- , negociation, and e s orysis; so z'c: organization can make constat small, adArtive :ri:43 :ather tht_ beias forced to undetrake massive L-ruptive ones. 4.3 This inuroduces our t.airå major dimension :, an-alytlcal frgamework: Time. The dilfmma for designers is z'e prblei of belanciag tLe lcn-term aeeds which venÅ to be heavily weig-ted toward social andi orgazat.ionai institutio -building, with the short-term needs fer laurC -ing a productive ac9iM: ad building a physical infrastrucrure. Annex Page 5 In addition, as we have seen above, a dynamic, adaptive approach to organization and management of project activities is needed. While there might be "appropriate" structures and processes for the project at different stages/phases of its development, this structure is continu- ously evolving out of the existing situation, so that the design of project organization is continuously linked to the ongoing flow of external events. 4.4 Consequently, while continuity of policies and procedures may be desirable from a project point of view, it is not realistic to assume that the environment will remain constant. In reality, the process of change and adjustment to change is likely to affect any project during its implementation. The emphasis on internal logic of design leads to a belief that events are subject to greater control than they really are. As a result, projet organizations tend to be overdesigned or, more accu- rately, overspecified; errors of design tend to be diagnosed as problems of management or implementation; and learning takes place primarily between phases of the project rather than within the project itself. This cannot but lead to lack of success in meeting the major project objectives of achieving not only physical input and output, but also developmental effect and impact.