POLICY RESEARCH WVXORKING PAPER 2 862 lenure, ulversity, ana ,ommitment (Community Participation tor Urban Service Provision C, _ L 17 T -11 307MK V. LIll Uwe Deichmann Mattias K. A. Lundberg Nazmul Chaudbuty Thie Wnrl BRink Development Research Group Infrastructure and Environment - June 2002 POLICY RESEARCH WORKING PAPER 2862 Abstract What factors influence community participation in the Lall, Deichmann, Lundberg, and Chaudhury use delivery of urban services? In particular, does security of household level survey data for Bangalore, India, to show tenure enhance the probability of participation as it that tenure security has a significant impact on the provides individuals with incentives to act collectively in willingness of residents to participate even when pursuit of a common objective? And are collective efforts neighborhoods are diverse in terms of their cultural less likely to succeed when there is a high degree of background and welfare status. Their findings suggest heterogeneity in culture or endowments among that participation is possible in heterogeneous community members? communities when it is a means to a common objective and not a goal by itself. This paper-a product of Infrastructure and Environment, Development Research Group-is part of a larger effort in the group to examine factors influencing urban development. Copies of the paper are available free from the World Bank, 1818 H Street NW, Washington, DC 20433. Please contactYasmin D'Souza; room MC2-622; telephone 202-473-1449; fix 202- 522-3230, email address ydsouza@worldbank.org. Policy Research Working Papers are also posted on the Web at http:/ /econwTorldbank.org. The authors may be contacted at slal!1@worldbank.org, udeichrnann@worldbank.org, mlundberg@worldbank.org, or nchaudhury@worldbank.org. June 2002. (32 pages) The Policy Research Working Paper Series disseminates the findings of work in progress to encourage the exchange of ideas about development issues. An objective ofthe series is to get the findings out quickly, even ifthe presentations are less than fully polished. The papers carry the names of the authors and should be cited accordingly. The findings, interpretations, and conclusions expressed in this paper are entirely those of the autbors. They do not necessarily represent the view of the World Bank, its Executive Directors, or the countries they represent. Produced by the Research Advisory Staff Te,,ure, ,s. D(;vers; and Cm *s; n,. Community Participation for Urban Service Provision Snomik V T/7)) Uwe Deichmann I 1- A Y.L, A 'IVUMat.s A%.. A1. LudberLUug Nazmul Chaudhury Development Research Group The World Bank, Washington DC 20433 JEL Classification: D10. D71. H42. R20 This paper is part of a larger program to find ways to irnprove urban management through the collection, use and public disclosure of spatially detailed information and analytic methods, funded by the UK DFID's Urban Knowledge Generation and TooLkits program and the World Bank Development Research Group. We thank Sonia Hanmmam, Christine Kessides, Vijayendra Rao, Ajay Suri, and participants of a seminar at the 14' Annual Bank Conference on Development Economics (2002) for useful comments. The findings repo.ted in 'uLS pap-er ae ULUse of uie -uu[S alone, adU bsnOUlU niU DU auib-OULU LU ine worlU Bans, its executive directors, or the countries they represent. Corresponding author: Somik V. Lall (cla*l rn--Aan' rg ), !81 Street T NW, V-OaAsTTtL, D 2A&.J' Tenure, Diversity and Commitment Co,munist Pariticnpations for L)rban V Pwori viPidnn L Introduction Rapid population growth in developing country cities creates an ever increasing demand for basic public services. City govermnents often do not have the capacity, know-how and fiscal resources to provide adequate services to newcomers while addressing the existing backlog in under-serviced areas. Public provision is therefore not always efficient and affordable, and sometimes does not consider the needs of individual communuties (iOsiom, IYY6). rnivate service pro-vision is atn UoLpti in soin llui WIlKVs. FrV x xiratav aiitwli, xupy by fsrker runtr or garbag pid1!e1- ,on is f ,wnA eyib .L %SR WV U'- CUjJjJFlJ LJ7 44l1l L -I -S r'W t5* **t hi f.A F. -J IV flW L pnrivate contractors in narts of some cities. However. the opportunities for cost recovery are often limited, especially in poorer communities. Consequently, private contractors frequently have limited incentives to provide basic services. In situations where government is unable and markets are unwilling to provide basic services, community based efforts may be able to fill the gap. Isham and Kakhonen (1999) find that collective action improves the performance of community based water supply projects in Indonesia, particularly as households gather collectively for operation and maintenance activities suich as cleaning drains and tanks. Pargal et al. (1999) evaluate determinants of community organization for garbage collection services in Dhaka, Bangladesh and find that honieowners have stronger community ties than those who are temporary residents. If seif-priovision of urban services by local communities and neignboUrniodU caU UV cUUniUdIVU a uel-ul ULnipLU11nL LU PUUIn ubi aIU pIVa4L pruvisioU, a ma,or rpoliyu uiPestioni is hnuo ri-mynivinih tl:v n cain eh tnimriraged and f2GilitqtPA bv governments and donors. To 'answer that question we need to know more about what motivates individuals to engage in community initiatives. iL1 LIndividuals may act. colectively fnr many reaeons= Tvnically; some degree of social cohesion is seen as a prerequisite. This issue has received renewed attention largely due to the work of Putnam (1995; see also Woolcock 1998, Knack 1999) who argues that community social interaction through clubs or other organized activities encourages the formation of 'social capital'. Social capital provides benefits to communities that are similar to agglomeration economies which provide benefits to firms. Proximity encourages face to face contacts and other types of interaction, which create knowledge spillovers and help develop norms of trust and reciprocity between agents which in turn reduce transaction costs (Collier i998, Fukuyama i995). ine main cirierence is maat cOMImuu[LLY pic LlJJ4LlII foI social LtU4iula%Atin is an exAInple VI a noVII. ur^tL be.-ef.t, wle naco1nm.Pmtihn1n ennnnmie es 2re. ey2mnpleq nf maricet henefits Bldiidla rld PlatL.eU (LYYJ) UdisItWInuI UVLWVVII HLLAUUU Ul l r,4 rU1t engageV Ili collectiven a-t.ionv bheicas thpey havA csir.ilar dndowmnity, cimilar r,ltura nr cimrilar objectives. Homogeneity in endowments and culture are important factors influencing participation in primarily social communities such as church groups or recreational clubs (Alesina and La Ferrara 2000). Economic benefits through networking at social functions or on the golf course will be a useful side product. However, the primary reason for joining is the ability to interact socially and 'bond' with other members of the community who tend to have similar incomes or share the same religious, ethnic, or linguistic background. It is not clear that these 'bonding' factors matter to the same extent when individuals act collectively in pursuit of similar objectives. The example we are concerned with in this paper is the self-provision of basic services such as access to water or solid waste collection. To achieve a specific task whose benefits are realized by each COLMIMLL CUj mebrindi-vidually, .-esidtIILs areV IkeIy to overcor.e dUIIlc i- e.ndowmen.ts or cultuhre. Here commulnity interaction is a means to anend, rather than an end in itself. The limited literature on this topic largely focuses on the impact of inequality on collective action. In a study of group formation in rural Tanzania, La rerrara (2000) mius LLIL miequality Lends to reduce membership in groups that nave economic objectives. The rich are mor,e likely to drop out in co-4nu 0tes kft. h r ineauahtv and individuals tend to sort into homogeneous income grouns leading to a decline in group functioning. Baland and Platteau (1997) develop a model to examine the impact of inequality on efficiency in the management of common property resources. While they find an ambiguous impact, they conclude that participation may decrease as inequality increases due to free rider problems in using common resources. In this paper we investigate what factors determine willingness to participate in community based initiatives when there is a homogeneity of objectives but not necessarily homogeneity in culture or endowments among community members. Controlling for community diversity, we are most interested in examining the role of security of tenure. There are several reasons why tenure status would contribute to enhanced participation in community programs for service provision. Tenure security proviUde ui%%.A.Uv* for iniviUda LV IIIVsbL in tUIhe kA3LLIMILLY UVbe4Ucas LUCe gUiUls 10111 imnmrnvem.nt in Qervices can hp. pnita1pi7ed in the hom vluep .o%ffand Sr '2A0A). Home owners will also expect to live in their residence for a long time and will thus anticipate an extended future stream of benefits from any improved service. In rigid, regulated housing markets (rent control, renters protection), some renters will expect to see the same long term benefits and therefore have the same incentives as owners. Home owners also tend to have lived in the community for a longer time. The longer the period of residence, the greater will be the level of social interaction between people and thus the community's social capital (Hoefferth and Iceland 1998). Using data for the US, DiPasquaie and Giaeser (i999) lind that home owners tend to be more engaged in coMmUniU y groups, letifg 'UIVL, LcorI,LIUUd Ui1t home nUmtuwi tiiu Lo oc betier citizens. This is a manor motivation for tax inrcentives arnd otfer programs th,at encouage home ownership in Western countries. The same is likely to hold in develoning counties. Using household level data from Ecuador, Lanjouw and Levy (2001) show that households with a stronger claim to their property are significantly more likely to 3 participate m acuviues to improve uie commumLy. Conversuly, B, U1 99l1773) shiuws that a large miant poplation does ,not facpi1itatP r,nnnpratinn in iirbnn conmmirniites. we use a recently compieted survey from Bangalore in soutuern Iniua to exainiime 'Ut Ulpd.L of1 LeLare scCu,.t o vJl ULhV wLIUJlrl.gl%OO ho uIseLLdsJP topar&Liate i commurnity based service deliverv programs= We are also interested in the auestion to what extent the shared obiective of improving urban services can help overcome differences in culture and social background in heterogeneous communities. In other words, to what extent does evidence in Western countries on the determinants of participation in social clubs hold when the focus is on community participation to achieve an economic objective in developing countries? A basic assumption motivating this question is that community based service provision is a valuable complement to other forms of service delivery. We do not discuss whether this generally holds true or in which circumstances community based service provision is more cost effective, provides beiter quaU4jLy IIILcaZLI orLU ia soc.Jaiz Jlyp ra"UI t J &JVV.ILU.d or FIVILL pJ.-o.sio. These arp imnnrtnint muestiPni nn .whinh evidenre sn far is limited. Tne remainder of uie paper is orgaiuzeu as iuluows. iII seLcIon ii, we uiscuss Uie aalyk app.-oach oani A-cn.be the fao,ars or. toh crn.munitr vi%uicipina,tion in servince deliverv. Section m describes the data and econometric specifications. We discuss results in Section IV, and present conclusions and policy implications in Section V. II A Model of Community Participation In this section we present a model of community participation in which the individual participates in order to achieve a certain goal such as an economic benefit or an improvement in living standards. Participation is thus not driven by the desire to obtain purely social benefits associated with belonging to a particular group. Also, the decision to participate is not primarily motivated by a concern for one's neighbors, but to obtain private, excludable benefits for oneself. Say for example tiat the individuai lives in an 4 area tfhat is uHnudr-serveu Dy municipal snrv-ices, suuii as wkterL. ri Allay LtLll udeiude tu ,oin f,tj hi iieid,hbrw to~ i-bt~i, better ivter iAjwdpi,Ai=nt1u of the city's rm^>amn supply. The model's self-interested individual could easily be allowed to care for his neighbors, but only to the extent that his neighbors' circumstances affect his own property values. Our model is based on two simple testable assumptions. First, the returns to participation are higher for owners than for renters. To the individual, there are two benefits to participation: it yields a flow of services, and the flow of services is capitalized in the value of the property. Running water is preferable to no running water, and a house with running water is more valuable than a house without. Presumably, the value of housing stock accrues to the owner, and not the renter. Formally, the household maximizes an additive, unitary inter temporal utility function of the flow of services and the consumption of other goods: Ut=u(s (pd xrc (d) (1) ~ he. e 've3i the %.o of SWAT ice fA. OfVA%0 p& L..C. ya ..AA.qISL, a..t s.rtWiS.' .0e i.& AArT o utility from the consumntion of other goods; less the cost of narticination The household maximizes utility subject to an inter temporal budget constraint, which in any period is defined as its previous wealth, plus any net changes to wealth, W = WWI+rd) + N(y t-x1-c(Dd) + h(pD (2) w.here-er is th.e peiod.a t interest rate, 'A. = x, = ct3 snet s,ntAr.g, nc ad h(p3 are +th rn*, -m to community participation that are canitalized in the stock of housing. This last benefit accrues primarily to the individual who has exchange rights to the property. If the rental market clears, and there is no oDnortunity for a renter to profit from "key money" or other sale of rental rights, the capital benefits to investment accrue solely to the owner. In distorted rental markets, such as Bangalore, some of these residual claims may accrue to ' The use the male pronoun is justified in this case -evidence reported below suggests that female-headed households are less willing to participate. the retr Tn ou. empiric n s+c-n Upl-n, we l-n, sor..e r-nterg to receive capita! U1I I ~IVILV4*III UL. VLL '.IALIIjJJ L L.Ui 0,V%,L%LJ.I U~LA J V", "AVY UL. Yr C AI L wI1W%L n w ~ benefits from nommmunity investments= We cLIn sUiVv U.hsV UU.ILY ILLUrLAIU LU UV'IVV, Uiv V1ULIIULILLI IVVlev UIop . ILi ivind Aiial % nrtticpnatfea if the benefits frnrn t,rtinnatinin (thln*) andi rn*)) nre areptpr thain the costs of participation (c(D*)). It is intuitively clear that the owner (or the renter with quasi-ownership rights), who receives both benefits, is more likely to participate for a given cost of participation. The equilibrium level ofp* will be a function of the cost of participation, and a vector of prices/costs associated with the flow of services from participation and household consumption. Lack of access can be due to two reasons: The household cannot afford to pay for the service, or the service is simply not available in a community either because of physical constraints (e.g., in the case of water supply, there may not be a connection to the trunk network) or because certain population groups are given priority in service access over others. Both of these points are related to the issue of community heterogeneity in endowments and in sociai background or culture as wili now be A.scuss-A ;..- I.. %Ac..eA "n -n Wealthier households tend to have better access to basic services. Tne rich can pay user cllarge or side 'pa-ymientbs LU obtain lri-i'Ve LV Uli JUU OL VILV. Ur p oUrI houlspholds, on th.e other hand, the cost of external service provision n ul represpntt a much larger share of their total disposable income. For that reason, they will be more inclined to explore less expensive alternatives such as self-help groups. They will also be more likely to contribute labor rather than money if their opportunity cost of time is lower than the potential charges. We therefore expect lower income households to be more inclined to participate in community efforts than wealthy residents. The poorest communities, on the other hand, may face different constraints to community participation. Squatter settlements, for example, are often inhabited by migrant laborers and recent immigrants. These communities may not have accumulated sufficient social 6 capital to enable community action.2 Taken together we expect the following ptern: if cost of servinces does not vnrvqtnn residents with tifferent welfare status_, pnrticipation will be lower for the poorest households, then initially rises with increasing welfare, and finally drop off for wealthier residents. At the community level, we might then expect households in areas characterized by larger variations in welfare status to be less willing to participate, if the success of community action depends on the participation of all or most members. This barrier to participation may be overcome, however, if the gains from achieving service improvements from participation will outweigh the cost associated with the increased risk of failure in economically diverse neighborhoods. Alternatively, services may be provided preferentially to a social group such as the members of an ethnic majority that has more political influence among local decision makers. In that case, those who are members of minorities will be less likely to receive pu-ulic sev-i. IAibInI, BiJ[jlr aUnU EaSLely I YY ( ) fInUU UliHL WV p[oVisloir Uo pUblic gUUUS is inversely related to ethnic hetero y, altoug, th-ey do not discuss the distribution of services within heterogeneous cities. This imnlies that ethnic minorities in diverse neighborhoods will be more likely to engage in "goal-oriented" (as opposed to merely social) participation than members of the majority. This leads to the second testable assumption. As discussed above, participation in social communities tends to increase with economic or cultural homogeneity. To put the argument crudely, individuals have an aversion to heterogeneity, if it implies that they will be forced to associate with people who are not like themselves. One "cost" of participation, therefore, is the act of association itself. This cost is higher in more diverse communities. If this is the case, we will see participation declining with heterogeneity. However, if the goal of participation is to obtain a necessary service that is not publicly provided to members of minority groups, then the cost of associating with non-alike perbon wile mrore thn4stbysebnfisrcie lllljiri- O-iluiz peoiauon wI I uu u U1all VIi0r.L Uy Uliv UV1IVILOIUb 1.I~VVU ILIU JUJLLUIIr 4 LAJIiIIIIU1Lil~Y 2In focus group interviews in Bangalore, residents of some of the poorest communities said that they would not engage in community activities because they do not trust their neighbors (Deichmann, Lall and Suri 2002). 7 iru.t;ative. in +1.at case s +.ht.-ogeity, will havt e no effect ot.nn par,+;c.nz-;n or maee inr.rease wiingness to narticinate to overcome some form of discrimination. '.JLLL MUIUUJ1 U1LVVIULU V,5LU11ALV,5 Mr. JlllliVIULU4 b UVIU1alUl Lu aU jJI%AjJL~ LL 4 VVOUj me a frntt.;on ofa vun2rnP ofinf;iu Aiigl andi rnrnmwn,tv r.harnertprctir. Frnom the discussion above, these community characteristics include measures of heterogeneitv in endowments and culture. These are generally presumed to decrease participation, but it is likely that homogeneity in objectives - the desire to obtain the service - will outweigh these effects. In addition, it is the members of the many heterogeneous minorities who are more likely to collectivize. Given that we do not explicitly have information on all the specific component of p*, we proxy the costs of participation and price variables with individual, household and community characteristics. We can express the structural equation underlying the observed behavior as Pi = a'W, + fi + ei (3) uwhee AP* is the individiiiP1' npt henefit fmn m6nrinsit.ingo W. is a vector of owm characteristics, H, is a vector of community or neighborhood characteristics that influence the individual's'decision, and e, is a normally-distributed error term with mean zero and variance a. We do not observe the latent variable Pi*. We see only the results of the individual's evaluation of (3), which is manifest in the choice made by the individual to participate or not to participate: D. = I iPD. ° () A dL~~J 11 ~~~1/ (JUJ Pi =O ifP¸ :vvv a micrnoronnnmtric anproach to development policy, Johns Hopkins University Press, Baltimore. Deaton, A. and S. Zaidi. (2000), Guidelines for constructing consumption aggregates for welfare analysis, Princeton University and Development Research Group, World Bank, Washington, D.C. T%e .~TT 0. W T ,-11 --..A A Q- )I (fT-C-.l 'bse i.A- ;-f, A IJiUlul"UL, U., S. V . Lall aWrI A. LiUL (20'02), A J1Li onWL Vi b s4L;.Ls- -S. --en management: survey design and findings from Bangalore, World Bank, Washington DC. Dietz, R. (2000), Estimation of neighborhood effects in the Social Sciences. An interdisciplinary literature review, Urban and Regional Analysis Initiative, Working Paper No. 00-03, Ohio State University. Fukuyama, F. (;995). Trusi; tie social virtues and the creation oj prosperity. Free Press, New York. Grosh, M. and P. Glewwe (2000), Designing household survey questionnaires for developing countries. Lessons from 15 years of Living Standard Measurement Survevs, World Bank. Washington, D.C. Isham, J. and S. Kahkonen (1999), What determines the effectiveness of commnunity-based water projects: Evidence from Central Java, Indonesia on demand, responsiveness, service rules and social capital, Social Capital Initiative Working Paper No. 14, World Bank, Washington DC. Jimenez, E. (1985). Urban squatting and community organization in developing countries, loum, ,,l& of P* U cor.omL.;c, 27, 69=92. Kahkonen, S. (1999), Does social capital matter in water and sanitation delivery: A review of the literature, Social Capital Initiative Working Paper no. 9, World Bank Washington D.C. Xerssi, C. (.VV02. IUUUWaoLL iV fAIIr.V UL UrUL .Iares as CVLLL'GI;I fUJ UUnIUVILULn aInU transformation, In World Development Report 2003, World Bank, Washington D.C. Knack, S. (1999), Social capital, growth and poverty: A survey of cross-country evidence, Social Capital Initiative Working Paper no. 7 World Bank, Washington D.C. La Ferrara, E. (forthcoming), Inequality and group participation: Theory and evidence from rural Tanzania, journai of Public Economics. Lanjouw, J.O. and P.I. Levy (2001), Untitled: A study of formal and informal property rights in urban Ecuador. Malpezzi, S. and V.K. lewari (1991), Costs and benefits of residential rent control in Bangalore, India, Infrastructure and Development Departnent Report INU 82, World Bank, Washington, D.C. Ostrom, E. (1996), (rossing the great divide co-pnroduction, synergy, and de-vte-lopnment, Worlld Development, 24, 6, 1073-1087. Pargal, S., M. Huq, and D. Gilligan, (1999), Social capital in solid waste management: Evidence from Dhaka, Bangladesh, Social Capital Initiative Working-Paper No. 16 World Bank, Washington DC. 23 Payne, G. (2001). Urban land tenure policy options: Titles or rights? Habitat International, 25, A14C AA Putnam, R. (1995), Bowling alone: America's declining social capital, Journal of Democracy, 6, January, 65-78. Whittington, D. (1998), Administering contingent valiation s-urveys in developing countries, World Development, 26, 1:21-30. Woolcock, M. (1998), Social capital and economic development: Toward a theoretical synthesis and policy framework. Theorv and Society, 27, 2:151-208. ;FIL ure 1: 11axr cdpad"0n dec,as.on ~ I 0 Households definitely -w Ding to partidpaf Yr< [W~'"~ ~ AJI other households 2 0 2 1 ( F 1 Wardboundaries 25 Figure 2: Per capita annual consumption quartiles First Second "~~~~~~~~~y- Third Fourh -1 ,,'1 '. r~~~ ~~~~~~~~~~~~ ~~t4~~t ¾, 4 26 Figure 3: Type of contribution by consumption quintiles 100% - < l j a 90% f _ _ _ _ E 80%J |- / NMoney- Inital investment 1 u 70% 1 \ -u-_Mney-O&M I cLabor- Inifial investment 9 7\I% /Z/\~ i |- Labor-O&M IL 40%- 30% 1 2 3 4 5 Per capni consumpton quintibes 27 Figiure 4: Distribution of major language groups in Bangalore Urdu Kannada 'k~ ~ ~~4 .x M . -. -r t V Tamil Telegu - .. . 28 Table t: Variahles Participation 1 Household responded that it would definitely participate in community initiatives to improve urban services Particination 2 Household resnonded that it would definitely or nrnbably participate in community initiatives to improve urban services Individual Tenure sft.tus H hOse ,nold o,,. d eW1iVTg , ,-t o.r 1T, hived ;n sune fo 20 ye or Housing category 1 = squatter settlement, non-notified or notified (base case) 4- XW MLUvUb dUve 3= Unauthorized revenue site or Vatarra 4= Government provided plots or flats, Cooperative or employer housing 5 = Private builders or City Improvement Trust Board Type of house 1 = free standing house 2 = attached house 3 = flat / apartment (base case) Welfare status Annual per capita household consumption (Rupees) Education Household head has at least a high school degree Age Experience - Age of household head Female HHH Female headed household Scheduled caste Household is a member of a "scheduled caste, tribe or other backward claSvi Language Household's mother tongue is the dominant language group in community Community Fr151ft,lItflSa * ProbYba.i. LuJUt .VT T husehot,ldsl in the, t,tnn&*z..t have Vt, same OGLI, *UILU tongue -,dLlguageu dor1u uc a S t--e LUC UUhedU1Lnu language among nouscnolus m the community Inequality Gini of household annual per capita consumption in the conmmunity Welfare level Average annual per capita household consumption of neighbors in the community (Rupees) Group homes initially built for industrial workers 29 Table 2: Variable means and standard deviations Mean Std. deviation Dependent variables Households definitely willing .746 .435 to participate Households definitely or .391 .488 probably willing to participate Individual variables T. enure status (Secure tenure) ArR2Q 0.4R3 Housing category 1 0.076 0.264 flUUSiII L~LCaU---- 1 0.0% 13 A. 4 413 Housing category 3 0.578 0.494 Housing category 4 0.164 0.370 Housing category 5 0.169 0.375 House type 1 .365 .481 House type 2 .600 .489 House type 3 .035 .183 WAef.rA rtatuis (in 10nOls) 3.068 2.086 - squared (in 10,000s) 13.797 31.979 Eduuailon .A r4 Age 47.35 12.34 Female headed household .075 .263 Scheduled caste .128 .335 Community defined as nearest Community defined by ward neighbors boundaries Mean Std. deviation Mean Std. deviation Fragmentation 0.609 0.139 0.617 0.129 Lang.1age rinminance 0.n32 0.156 0.531 0.144 Community welfare level 3.034 0.837 3.068 0.901 -squared 9.911 5.732 10.226 6.949 Inequality 0.265 0.058 0.269 0.063 TableU 3: v.ruuuea io pas uczpuab ard iype of cou.botlor. (perer.. ,wo, old you PBrc.P,nt Tun,- nf T,ibhial (ra2hinine &r participate? contribution investment Maintenance Definitely yes 39.1 Money 77.8 48.5 Probably yes 35.5 Labor 49.1 67.4 Probably no 16.4 Definitely no 9.1 30 Table 4: Estimation results for households definitely intending to participate Community defined as nearest Community defined by ward neighbors boundaries (1) (2) (3) (4) Individual variables Tenure status 0.125 0.125 0.123 0.123 (0.019)** (0.019)* (0.019)** (0.019)** Housing category 2 -0.063 -0.065 -0.072 -0 073 (0.088) (0.088) (0.087) (0.087) Housingcategory3 0.122 0.122 0.115 0.114 (0.038)** (0.038)* (0.038)* (0.038)** Housina cateaorv4 0.22 0 219 0.213 0.214 (0.045)** (0.046)* (0.046)* (0.046)* Hnining categonry 5 0.108 0.107 0.103 0.102 (0.045)* (0.045)* (0.045)* (0.045)* HnI ieptvnA 1 nn011 n0n12 n.nn0 n.nn (0.054) (0.054) (0.054) (0.054) House tyne 2 0.022 0.023 0.019 n.n2 (0.053) (0.053) (0.053) (0.053) Welfare status 0.023 0.023 0.028 0.028 (0.010)* (0.010)* (0.010)** (0.010)** snsuared -0.0011 -0.001 .V0V01 =n0n (0.001)* (0.001)* (0.001)* (0.001)* EduCation 0.057 0.056 0.053 0.053 (0.023)* (0.023)* (0.023)* (0.023)* An,= flArg nnnna A Age 0.05 0 05 0 05%0.Il (0.031) (0.031) (0.031) (0.031) Fenma!e headed household n.naa -0.n08 -n.nn2 n nn- (0.034)* (0.034)* (0.034)** (0.034)** nd44ule + 0 04112 0.14A 4.11 (0.029)** (0.029)* (0.029)** (0.029)** Lnguage -n.n036 -10.033 ^-0.010.1 (0.019)+ (0.020)+ (0.019) (0.019) ComMoU..^vrfb Fragmentation 0.123 0.114 (0.075)+ (0.080) Language dominance -0.103 -0.136 In A&& IA A-%A Community welfare level 0.095 0.094 0.142 0.135 (0.069F) (0.0o6) (0.02)* (.062)* - squared -0.016 -0.016 -0.023 -0.022 [v) I0)+ (0.010) (0.007)** (0.007)** Inequality 0.261 0.263 0.171 0.212 (V.1 76) (0.175) (0.172) (0.i69) Observations 2901 2901 2901 2901 F-tests for joint significance Welfare status & -squared (5.08)+ (5.04)+ (7.18)* (7.36)* Welfare level & -squared (4.45) (4.66)+ (18.55)** (18.84)** Fragmentation & language (8.21)* (8.43)* (5.021+ (3.54) Table 5: Estimation results for households definitely or probably intending to participate Community defined as nearest Community defined by ward neighbors boundaries (1) (2) (3) (4) Individual variables Tenure status 0.12 0.12 0.119 0.119 (0.018)** (0.018)** (0.018)** (0.018)- Housing category 2 -0.142 -0.143 -0.138 -0.139 (0.086)+ (0.087)+ (0.086) (0.086) Housing category 3 0.073 0.073 0.077 0.076 (0.033)* (0.033)* (0.033)* (0.033)* Housing category 4 0.13 0.13 0.132 0.132 (0.029)** (0.029)** (0.029)* (0.029)- Housing category 5 0.044 0.044 0.05 0.049 (0.034) (0.034) (0.033) (0.033) Housing type 1 0.087 0.088 0.09 0.089 (0.044)* (0.044)* (0.044)* (0.044)* Housing type 2 0.029 0.03 0.032 0.033 (0.047) (0.047) (0.047) (0.047) Welfare status -0.016 -0.016 -0.016 -0.016 (0.010)+ (0.010)+ (0.010) (0.010) - squared 0.001 0.001 0.001 0.001 (0.001) (0.001)+ (0.001) (0.001) Education 0.044 0.044 0.041 0.041 (0.021)* (0.021)* (0.021)+ (0.021)+ Age 0.003 0.003 -0.001 -0.001 (0.029) (0.029) (0.029) (0.029) Female headed household -0.01 -0.01 -0.013 -0.013 (0.032) (0.032) (0.032) (0.032) Scheduled caste 0.08 0.08 0.079 0.079 (0.022)** (0.022)** (0.022)** (0.022)** Language -0.021 -0.021 -0.014 -0.017 (0-017) (0.018) (0.017) (0-017) Community level variables Fragmentation 0069 0 109 (0.065) (0.069) Languae dominne -0.082 -01 (0.057) (0.064)* f'^mm Iniftu .AIIIfare level 0.003 0.003 0.06 0.053 (0.060) (0.060) (0.043) (0.043) -squared -0.004 -0.004 -0.01 -0.01 (0.009) (0.009) (0.005)+ (0.005)+ lnequaliby n.n103 -0.009 0.033 (0.157) (0.156) (0.152) (0.149) Observations 2901 2901 2901 2901 F-tests for joint significance Welfare status & -squared (2.89) (2.94) (2.71) (2.55) Welfare level & -squared (3.69) (4.14) (7.84)* (8.26)* Fragmentation & language (4.78)+ (3.90) (6.24)* (4.49) WPS2837 Reform, Growth, and Poverty David Dollar May 2002 E. Khine in Vietnam 37471 WPS2838 Economic Mobility in Vietnam Paul Glewwe May 2002 E. Khine in the 1990s Phong Nguyen 37471 WPS2839 Marketing Externalities and M. Shahe Emran May 2002 F. Shilpi Mv . arket rDevelopment Forlhad Sh;i'pi 87476 WPS2840 Public- Spnnding and iuitcnmsp: Andrew Sunii Rqikiimar Mav 2002 H. Sladovich Does Governance Matter? Vinaya Swaroop 37698 WPS2841 Contractual Savings in Countries Gregorio Impavido May 2002 P. Braxton with a Small Financial Sector Alberto R. Musalem 32720 Dimitri Vittas WPS2842 Financial Sector inefficiencies Pierre-Richard Ag6nor May 2002 M. Gosiengfiao and the Debt Laffer Curve Joshua Aizenman 33363 WPS2843 A Practical Guide to Managing David Scott May 2002 L. Yeargin Systemic Fin:ancial Crics:c A RPview R1S5 of Approaches Taken in Indonesia, the Republic of Korea, and Thailand WPS2844 Money Demand in Venezuela: Mario A. Cuevas May 2002 M. Geller Multiple Cycle Extraction in a 85155 Cointegration Framework WPS2845 The Spatial Division of Labor Marcel Fafchamps May 2002 F. Shilpi in Nepal Forhad Shi,pi 87476 WPS2846 Es indina'cs Eonnomic' !r Alth I eavinq rGairax natt My 202 G. Cuiinanan the Poor Behind? Martin Ravallion 32301 WPS2847 The Nature and Dynamics of Poverty Hippolyte Fofack May 2002 P White Determinants in Burkina Faso 81131 in the 1990s WPS2848 Administrative Barriers to Foreign Jacques Morisset May 2002 M. Fegnai Investment in Developing Countries Olivier Lumenga Neso 36177 WPS2849 Pooling, Savings, and Prevention: Truman G. Packard May 2002 T. Packard Mitigatinn the Pisk of O!dr4 Ana 7841 Poverty in Chile WPS2850 Determinants of Commercial Bank David A. Grigorian June 2002 S. Torres Performance in Transition: An Vlad Manole 39012 Application of Data Envelopment Analysis WPS2851 Economic Development and the Bernard Hoekman June 2002 P Flewitt vvurld Trade Organizdato AIl-te UUldo WPS2852 Regional Agneements and Trade in Aaditya MAnttonn II ne 2002n P. Flewitt Services: Policy Issues Carsteri Fink 32724 WPS2853 Private Interhousehold Transfers in Donald Cox June 2002 E. Khine Vietnam in the Early and Late 1990s 37471 WPS2854 Rich and Powerful? Subjective Michael Lokshin June 2002 C. Cunanan Power and Welfare in Russia Martin Ravallion 32301 WPS2855 Financial Crises, Financial Luc Laeven June 2002 R. Vo Dependence, and Industry Growth Daniela Klingebiel 33722 Randy Kroszner vvPrS2o5u Ban'king Policy and IVMcro eVolnul lull Ilk, Gearld Claprio, JI. JVun IV A . Yzptenr o Stability: An Exploration Patrick Honohan 31823 WPS2857 Markups, Returns to Scale, and Hiau Looi Kee June 2002 M. Kasilag Productivity: A Case Study of 39081 Singapore's Manufacturing Sector WPS2858 The State of Corporate Governance: Olivier Fremond June 2002 G. Gorospe Experience from Country Mierta Capaul 32623 Assessments WP S2 8 59 Ethnic and' Gend'er WVage MohIlamedu ihlsan AJwad June_u_ 2002 Lileta Disparities in Sri Lanka Pradeep Kurukulasuriya 84321 WPS2860 Privatization in Competitive Sectors: Sunita Kikeri June 2002 R. Bartolome The Record to Date John Nellis 35703 WPS2861 Trade-Related Technology Diffusion Maurice Schiff June 2002 M. Kasilag and the Dynamics of North-South Yanling Wang 39081 and South-South Integration Marcelo Olarreaga