urban households' willingness to pay for improved solid waste

Hindawi Publishing CorporationUrban Studies ResearchVolume 2013 Article ID 659425 8 pageshttpdxdoiorg1011552013659425

Research ArticleUrban Householdsrsquo Willingness to Pay for Improved Solid WasteDisposal Services in Kumasi Metropolis Ghana

Dadson Awunyo-Vitor1 Shaibu Ishak2 and Godfred Seidu Jasaw3

1 Department of Agricultural Economics Kwame Nkrumah University of Science and TechnologyPO Box UP 1007 KNUST Kumasi Ghana

2 Kwadaso Agricultural College Ministry of Food and Agriculture Academy Post Office Kwadaso Kumasi Ghana3Department of Community Development Faculty of Planning and Land Management University for Development StudiesPO Box 3 Wa Campus Ghana

Correspondence should be addressed to Dadson Awunyo-Vitor awunyovitoryahoocouk

Received 28 November 2012 Accepted 13 March 2013

Academic Editor Cesar Ducruet

Copyright copy 2013 Dadson Awunyo-Vitor et al This is an open access article distributed under the Creative Commons AttributionLicense which permits unrestricted use distribution and reproduction in any medium provided the original work is properlycited

Solid waste management within Kumasi Metropolitan Assembly area continues to be a major challenge for the municipal assemblyand one of the key issues is its financial constraints This study was undertaken to examine householdsrsquo willingness to pay forimproved solid waste management services A multistage sampling technique was employed to select six hundred respondentsfor the study Logistic regression model was used to establish the determinants of willingness to pay for solid waste managementwhilst the Tobit model was used to evaluate the factors influencing the amount of money the households are willing to pay forimproved solid waste management The logistic model shows that income age number of children quantity of waste generatedand education have significant effects on the willingness to pay while the amount of money the households are willing to pay wasinfluenced by their income quantity of waste generated education house ownership and number of children Thus the assemblycan increase waste collection fees between GHC 3 andGHC 500This would lead to improvement in the waste management withinthe metropolis However the additional charge should take into consideration location and income levels

1 Introduction

Waste is directly linked to human development both techno-logically and sociallyThe composition of different wastes hasvaried over time and location with industrial developmentand innovation being directly linked to waste materialsSome components of waste have economic value and can berecycled once correctly recovered

Humans generate a great deal of waste as a by-product oftheir existence This is evidenced at dumping pits located inor around archaeological sites Every task from preparing ameal to manufacturing a computer and so forth is accompa-nied with production of waste material which cannot be usedfor other things and needs to be disposed of effectively If notcontained and handled appropriately waste can balloon intoa huge problem as for example when garbage ends up in theopen oceanwhere it canmake animals and birds sick amongstothers

Waste is sometimes a subjective concept because itemsthat some people discardmay have value to others It is widelyrecognized that wastematerials are a valuable resource whilstthere is a debate as to how this value is best realized Suchconcepts are colloquially expressed in the western culture byidioms like ldquoOne manrsquos trash is another manrsquos treasurerdquo Onthe generation end waste management agencies have placedan increasing focus on reducing waste so that there is lessto cope with This can be done on an industrial level bydevelopingmore efficient processes reducing packaging andso forth Individuals consumers can also make commitmentto generate less waste A big part of this movement hasfocused on recycling in which usable goods are reclaimed sothat they can be reused or repurposed

Transportation of waste is a major issue as appropriatedisposal sites may be remote Frequently subscription pick-up services are available for people paying a flat fee to havetheir waste picked up and disposed of Other people can

2 Urban Studies Research

also subscribe to specialty services like medical waste pick-up services or confidential paper shredding and disposalservices

Waste management practices differ for developed anddeveloping nations for urban and rural areas and for resi-dential and industrial producers For instance in some casesmanagement for nonhazardous residential and institutionalwaste in metropolitan areas is usually the responsibility oflocal government authorities while management for hazard-ous commercial and industrial waste is usually the respon-sibility of the generator Developing effective waste manage-ment strategies is critical for nations all over the world asmany forms of waste can develop into a major problem whenthey are not handled properly Numerous firms provide wastemanagement services of a variety of types and several gov-ernments also regulate the waste management industry forsafety and efficacy

According to [1] historically the amount of waste gener-ated by human population in the early ages was insignificantmainly due to the low population densities coupled with thefact that there was very little exploitation of natural resourcesCommon wastes produced during the early ages were mainlyashes and human and biodegradable wastes and these werereleased back into the ground locally with minimal adverseenvironmental impact

Before the widespread use of metals wood was widelyused for most applications However reuse of wood has beenwell documented Nevertheless reuse and recovery of suchmetals have been carried out by earlier humans With theadvent of industrial revolution waste management became acritical issue This was due to the increase in population andthemassive migration of people to industrial towns and citiesfrom rural areas There was a consequent increase in indus-trial and domestic wastes posing threat to human health andenvironment

In Africa municipal solid waste management constitutesone of the most crucial health and environmental problemsfacing governments of African cities This is because eventhough these cities are using 20ndash50 percent of their budgetin solid waste management only 20ndash80 percent of the wasteis collected The uncollected or illegally dumped wastes con-stitute a disaster for human health and the environmentaldegradation Not only is their quantities increasing but theirvariety is also increasing both a consequence of increasingurbanization incomes and changing consumption habitsfuelled by globalizationThis scenario places the already des-perate urban councils in a difficult situation especially as theyhave to develop new strategies to deal with increasing vol-umes as well as strange varieties of wastes Poor waste man-agement practices in particular widespread dumping ofwaste in water bodies and uncontrolled dump sites aggra-vates the problems of generally low sanitation levels acrossthe African continent

According to [2] urbanization is on the rise inAfrica andthis trend is expected to continue in the future Of concern isthe inability of infrastructure and land use planningmethods(including waste management) to cope with urban growth(the highest in the world) at 35 percent annuallyThis is part-icularly urgent in slum areas which constitute a big part of

many of the cities and towns in Africa Waste managementinfrastructure is largely nonexistent in rural areas of Africa

Imports of second-hand consumer goods and productionandor import of substandard products are all contributing tothe rapid increase in waste generation Policies should be putin place and existing standards enforced to reverse this trendImplementation or enforcement of waste regulations andconventions is severely constrained by the lack of good gov-ernance transparency and prevalence of corruption in somecases Lack of awareness and appreciation of best practicesfor environmentally sound management of wastes is a majorconstraint A paradigm shift among communities and societyat large is needed Again the fast growing use of ICT andrapid turnover in technology (particularly computers mobilephones etc) is creating a growing E-waste stream for whichthere is no waste management capacity yet This leads to dis-posal of both E-waste and municipal waste in dump sitesChanging lifestyles and consumption patterns of the growingurban middle class in particular increases the complexityand composition of waste streams in Africa

The gap between waste management policy and legisla-tion and actual waste management practices is widening dueto perennial capacity constraints and lack of waste manage-ment facilities for various waste streams Access to majorinvestments and acquiring the technical know-how neededto resolve the capacity constraints remain a tall order Wastegeneration is expected to increase significantly as a result ofindustrialization urbanization and modernization of agri-culture in Africa This will further aggravate the current cap-acity constraints in waste management

Progress has been made in waste management policiesand strategies Biogas and compost production from organicwaste fractionation has been widely accepted in Africa as abest practice and progress is being made in developing andimplementing specific projects in various countries How-ever the use of economic instruments and implementationof polluter-pays principles in waste management have yet tomature in most African countries

The single largest implementation challenge formanagingwaste policies remains creating sufficient capacity for envi-ronmentally sound management including where appro-priate recovery and recycling of various waste streams acrossAfrica The effort to do this is constrained by access tofinance and technical know-how Current bylaws in mostAfrican countries give responsibility for waste managementto municipalities which are often ill-equipped to deal withcollection and disposal Such bylaws are now an impedimentto investment in waste management by the private sector

2 Problem Statement

Ghana in general andKumasiMetropolitanAssembly (KMA)in particular have made several attempts at addressing thewaste menace which is on the rise as a result of populationhikes growth in industrialization and consumer attitudesThe cityrsquos bylaws and policies on waste and sanitation seekto address the waste challenge in its entirety on individualbasis The assembly among its efforts and strategies has con-tracted some waste companies to handle waste collection and

Urban Studies Research 3

disposal It has also been implementing the polluter-paysprinciple to get individuals to pay for waste management ser-vices Furthermore some effort is being made to educate thepublic and create some level of awareness to enable membersof the public to play a role in reducing waste and handlingwaste efficiently

However the level of achievement of this objective leavesmuch to be desired as there is a presence of piles of wastes onthe streets market centres and homes Waste managementstill remains a herculean task to the Assembly as it has notbeen able to manage and deal with waste problem to theexpected level of it This situation according to the KMA isgenerally attributed to inadequate finance to clear the solidwaste in the case of both the assembly and contracted com-panies Thus the assumption is that if households pay morethen the services would be improved However very little hasbeen done to assess the householdsrsquo willingness to pay forimproved waste management services The question then isthat Are the households ready to pay more How much arethey prepared to pay andWhat factors determine theirmoti-vation to pay and the amount of money they are willing topay The main objective of this study is to assess the deter-minants of householdsrsquo willingness to pay for improved solidwaste management services and the amount of money theyare willing to pay

3 Methodology

31 Data Collection Procedure The sample for the study wasselected in three stages first was the purposive selectionof submetros followed by the random selection of electoralareas within the selected submetrosThe selection of the sub-metro was guided by the level of waste management activitieswithin these areas using a report from KMA [3] The thirdstage involves random selection of households In all 600households were selected for the study Data and informationwere collected through individual interviews using well-structured questionnaires

32 TheTheoretical and Analytical Framework Two levels ofanalyses were carried out The first was to estimate the deter-minants of household heads willingness to pay for improvedwastemanagement services using the logitmodelThe secondlevel of analysis was to estimate the determinants of theamount of money they are willing to pay for improved wastemanagement services using the Tobit model

The logit model as stated by [4] earlier was employed toexamine the determinants of household heads willingness topay for improvedwastemanagement servicesThe study usedthe threshold decision-making theory proposed by [5 6] toanalyse the determinants of willingness to pay for improvedwaste management services by household heads The theorypoints out the fact that when the individual is faced with asituation to take a decision in this case to pay for improvedwaste management services or not to pay heshe has areaction threshold which is dependent on a certain set offactors As such at a certain value of stimulus below the thre-shold no reaction is observed while at the critical threshold

value a reaction is stimulated Such phenomena are generallymodelled using the relationship

119884119894= 120573119883119894+ 120583119894 (1)

where 119884119894is equal to one when a choice is made to pay for

improved waste management services and zero otherwisethis means

119884119894= 1 if119883

119894is greater than or equal to a critical value

119883lowast and

119884119894= 0 if119883

119894is less than a critical value119883lowast

Note that 119883lowast represents the threshold value of the inde-pendent variables (119883) Equation (1) represents a binary choicemodel involving the estimation of the probability of willing-ness to pay for improved waste management services (119884) asa function of independent variables (119883) Mathematically thisis represented as

Prob (119884119894= 1) = 119865 (120573

1015840119883119894)

Prob (119884119894= 0) = 1 minus 119865 (120573

1015840119883119894)

(2)

where 119884119894is the observed response for the 119894th observation of

the response variable 119884 This means that 119884119894= 1 for a house-

hold head who is willing to pay for improved waste man-agement services and 119884

119894= 0 for a household head who is

not willing to pay for improved waste management services119883119894is a set of independent variables such as literacy monthly

income age marital status housing arrangement and quan-tity of waste generated gender associated with the 119894thindividual which determine the probability of willing to payfor improved waste management services (119875)The function 119865may take the form of a normal logistic or probability func-tion The logit model uses a logistic cumulative distributivefunction to estimate 119875 as follows

119875 (119884 = 1) =1198901205731015840119883

1 + 1198901015840119883

119875 (119884 = 0) = 1 minus1198901205731015840119883

1 + 1198901205731015840119883=

1

1198901205731015840119883

(3)

According to [7] the model is a regression of the conditionalexpectation of 119884 on119883 giving

119864(119884

119883) = 1 [119865 (120573

1015840119883)] + 0 [1 minus 119865 (120573

1015840119883)] = 119865 (120573

1015840119883) (4)

Since themodel is nonlinear the parameters are not necessar-ily the marginal effects of the various independent variablesThe relative effect of each of the independent variables on theprobability of a household head willing to pay for improvedwaste management services is obtained by differentiating (4)with respect to119883

119894resulting in (5) [7] as

120597119875119894

120597119883119894

= [1205821205731015840119883

1 + 1205821205731015840

119883]120573 = 119865 (120573

1015840119883) [1 minus 119865 (120573

1015840119883)] 120573 (5)


The maximum likelihood method was used to estimate theparameters The implication for applying the logit model inthis paper is that the respondents would decide to pay forimproved waste management services when the combinedeffects of certain factors exceed the inherent resistance of notto pay for improved waste management services The pre-ference for the logistic regression model to the conventionallinear probability regression model in analysing the deter-minants of household heads willingness to pay for improvedwaste management services is based on the fact that the para-meter estimates from the former are asymptotically consis-tent and efficient The estimation procedure employed alsoresolves the problem of heteroscedasticity and constrains theconditional probability of making the decision to pay forimproved waste management services lie between zero andone The main reason for choosing the logit model over theprobit model for this paper is because of its mathematicalconvenience and simplicity [6] and the fact that it has beenapplied in similar studies by [8 9] among others

The logit model provides information only with respectto the household headsrsquo decision to pay for improved wastemanagement services or not to pay but not on the amount ofmoney they arewilling to pay To estimate the determinants ofthe amount of money they are willing to pay the Tobit modelis employed The Tobit model allows us to identify the fac-tors that determine how much the respondents are willingto pay for improved waste management services The Tobitmodel was developed by Tobin in 1958 and has been used bya number of researchers including [10ndash12] in various studiesAccording to [6 13] the general formulation of the Tobitmodel is usually given in terms of an index function This isgiven in (6) as

119910119894= 1198831015840120573 + 120576119894 (6)

where 119910119894is the dependent variable in this case is the amount

of money the respondents are willing to pay 119883119894is a set of

explanatory variables and 120576119894is assumed to be an independ-

ently and normally distributed stochastic term with zeromean (120583) and constant variance (1205902) Assume that there is aperceived utility119880(119910) for paying for improvedwastemanage-ment services and a utility119880(0) for not paying for improvedwaste management services Further assume that there is acluster of the population with no decision tomake at the limit[10ndash12 14] then

119910119894= 0 if 119910lowast

119894le 0 for not paying for improved waste

management services119910119894= 1 if 119910lowast

119894gt 0 for paying for improved waste man-

agement services

where 119910lowast119894is the unobserved latent variable or the threshold

which is observed only when 119910119894or the amount of money

households are willing to pay is positive The expected value119864119910of the amount of money they are willing to pay for

improved waste management services is given as follows

119864119910= 119883119894120573119865 (119911) + 120590119891 (119911) (7)

where 119883 is the vector of explanatory variables 119865(119911) is thecumulative normal distribution of 119911 119891(119911) is the value of the

derivative of the normal curve at a given point (ie the unitnormal distribution) 119911 is given as119883120573120590 120573 is a vector of Tobitmaximum likelihood estimates 120590 is the standard error of themodel The relationship between the expected value of allobservations 119864

119910 and the expected conditional value above

the limit 119864lowast119910is given by

119864119910= 119865 (119911) 119864

lowast

119910 (8)

Analyzing the policy implications of changes in the relevantexplanatory variables is a major component of this paper Tothis end the effect of the change in 119894th variable of119883 on119910 leadsto the following decomposition

120575119864119910

120575119883119894

= 119865 (119911) (

120575119864lowast

119910

120575119883119894

) + 119864lowast

119910(119865 (119911)

120575119883119894

) (9)

Thus (9) suggests that the total change in elasticity of 119910119894

can be disaggregated into two parts namely the change inprobability of the expected level of intensity and the changein the elasticity of willing to pay for improved waste man-agement services

To obtain themarginal effect of the observed variable thatis of interest in this paper the following formula [6] is used

120575119864 (119910119883119894)

120575119883119894

= 120573lowastProb (0 lt 119910lowast lt 1) (10)

According to [7] the log likelihood of the Tobit model is spe-cified as

ln 119871 = sum

119910119894gt0

minus1

2

[

[

log (2120587) + ln1205902 +(119910119894minus 1198831015840

119894120573)2

1205902]

]

+ sum

119910119894=0

ln[1 minus 120601 (119883

1015840

119894120573)

120590]

(11)

Maximising this likelihood function with respect to 120573 and 120590gives the maximum likelihood estimates of these parametersStata (version 10)was the statistical software employed to esti-mate the empirical parameters by MLE

4 Choice of Variables for Logit andTobit Models

The variables used in the logit and the Tobit models are pre-sented in Table 1 The choice of variables was based more onrelated studies by researchers as follows

(i) Income This variable refers to the monthly moneyincome of the household in terms of Ghana CedisIt includes the income of the head from all sourcesThere is a general agreement in the environmen-tal economics literature on the positive relationshipbetween income and demand for improvement inenvironmental quality [15] Therefore we expect theincome to affect the willingness to pay and its amountpositively


Table 1 Description of explanatory variables used in the model

Variable Description Unit of measure

Income Monthly income ofthe household Ghana Cedis (GHC)

Gender Gender ofrespondents

Binary = 1 if male0 = otherwise

Age Age of respondents Years

Education Number of years offormal education Years

Marital status Marital status ofrespondent

D = 1 if married0 = otherwise

Time spent in thearea

How long respondenthas been living in thearea

Years

Housingarrangement

Housingarrangement

Binary = 1 if owneroccupied 0 = otherwise

Number ofdependents

Number ofhousehold membersbelow 18 years of age

Number individuals

Quantity of waste Volume of wastegenerated

Number of GHC 30worth plastic(polythene) bag

Responsibility ofsolid wastemanagement


Binary 1 = if they thinkKMA is responsible0 = otherwise

(ii) Gender This study expects female respondents to bemore willing to pay than men since traditionally it isthe role of women to clean the house and dispose ofthe waste

(iii) Age This refers to the age of the respondent and itis expected to affect the willingness to pay negativelyThis is because old people may consider waste collec-tion as the government responsibility and could beless willing to pay for it while the younger generationmight be more familiar with cost sharing and hencemay be willing to pay more for improved waste man-agement

(iv) Education This variable is taken to capture the num-ber of years the respondent spent in formal school sys-tem Education is expected to have positive and signi-ficant effect on waste management Thus the longerperiod the individual spent in formal school systemthe more likely that heshe would be willing to paymore for improved waste management

(v) Marital status The marital status of the householdhead is expected to influence the value the individualplaces on waste management This is due to the factthat married people are likely to be more responsibleto keep the environment clean and hence are morelikely to be willing to pay more for improved wastemanagement

(vi) Length of stay This refers to the number of years thehousehold has been living in the areaThis is expectedto influence the willingness to pay in the positive

direction since the longer the year the household hasbeen there the more they would understand the pro-blem of solid waste management of that area and themore they would be willing to pay for improvementin the waste management

(vii) Number of children in the household This refers tothe number of household members who are below 15years of age This variable is expected to have a neg-ative effect on the willingness to payThis is due to thefact that themore children in the household themorethey would prefer to use their children to clean theenvironment than paying more to the metropolitanauthorities to clean the environment

(viii) Quantity of waste generated This variable stands forthe quantity of waste the household generates withina week For the purpose of this study the unit of mea-surement used is a shopping plastic (polythene) bag(30 Ghana pesewas worth) which is common as aconvenient means for measurement to most respon-dents during the survey The study hypothesizes thewillingness to pay to be positively related with thequantity of solid waste generated since the higher thegeneration the more would be the problem house-holdsrsquo face in storage and taking the waste for collec-tion and they would be willing to pay more

(ix) Responsibility for solid waste management This vari-able is taken as a proxy to examine the attitude ofthe respondents towards who should manage wastein the metropolis This study expects positive attitudetowards cost sharing to influence the willingness topay in the positive direction It is specified as dummyvariable 1 if they think KMA is responsible for wastedisposal and 0 otherwise

(x) Tenancyhousing arrangement Those living in theirown houses are expected to be more willing to pay forthe improvement as compared to their tenants Thisis because the house belongs to the owners and if theplace is clean they may have a higher value for theirproperties

5 Results and Discussion

51 Willingness to Pay by Those Not Currently Paying Thestudy revealed that good proportion of 342 (57) householdsare willing to pay for improved services while 258 (43) areunwilling to pay for improved services Some of the reasonsgiven for the unwillingness to pay include the following

(i) There is nowastemanagement service provided in thearea

(ii) Some do dispose their waste in secondary receptacleof which they were not charged for

(iii) They dispose their waste generated in holes dugaround their homes

(iv) It is the responsibility of the government to pay forthem


Table 2 Logit regression results of factors influencing willingness to pay for improved waste management services

Independent variables Coefficient Std errors 119875 values Marginal effectEducation minus00154 00083lowast 0062 148039Length of stay minus00093 00032lowastlowastlowast 0003 116667Bags of waste generated 00024 00021 0255 179608Age 00029 00024 0236 482157Housing arrangement 01630 00812lowastlowast 0045 06470Distance minus00003 00002lowast 0077 07135Gender minus00975 00410lowastlowast 0018 08235Total income minus00003 00097 0975 64902Satisfaction minus00101 00445 0821 03333Responsibility minus00233 00355 0513 27255

Goodness of fit measureslog likelihood minus390401Restr Log likelihood minus1086294McFadden 119877-squared 06406LR statistic (14 df) 2391788lowastlowastlowastlowastlowastlowastSignificant at 1 lowastlowastsignificant at 5 lowastsignificant at 10

(v) It is not necessary to pay for waste when there areother equally important issues

Considering the above reasons education is important toencourage individual households to pay for improved wastemanagement services Also KMA should do more to ensurethat all submetro have their fair share of waste managementservices More households could be made to accept to pay forwaste services when the necessary conditions are brought tobear

52 Determinants of Householdsrsquo Willingness to Pay for Im-proved Waste Management The logit regression results offactors influencing willingness to pay for improved wastemanagement are presented in Table 2 The logit regressiongave aMcFadden 119877-squared of about 064The log likelihoodratio (LR) statistic is significant at one percent meaning thatat least one of the variables has coefficient different from zeroTherefore it can be concluded that the logit model used hasintegrity and is appropriateThe validity of the logit model inestimating willingness to pay for improved waste disposal isconsistent with related studies [16]

The number of years of schooling shows positive and sig-nificant relationship with the willingness to pay for improvedwaste management servicesThis result supports the findingsof [15] The higher the educational level the higher the pro-bability of the personrsquos willingness to pay for improved wastemanagement services This is because the fact that as indi-viduals receive education they tend to understand the needfor waste management better The marginal effect revealedthat an additional year of schooling would increase the like-lihood of personrsquos willingness to pay for improvedwasteman-agement services by about 15

Length of stay in the area variable is positive and signif-icant at 5 level of significance This is because the longerpeople stay in an area the more they are concerned about theenvironment and sanitation in the area An additional year

stay in an area within KMA increases the likelihood of indi-vidual willingness to pay for improved waste managementservices by 1166

The coefficient of housing arrangement variable is sig-nificant at 5 level of significance This result indicates thatlandlords are more willing to pay for improved waste man-agement services as compared to tenants This is particularlyso because in Ghana landlords are summons in case thecity authorities have problem with the sanitation and notthe tenants As the distance to dumping site increases thelikelihood of the respondentsrsquo willingness to pay for improvedwastemanagement services is higherThis is because increasein distance increases the cost (finance and time) of wastedisposal by individuals One kilometre increase in distance todumping site increases the likelihood of individualrsquos willing-ness to pay for improved waste management services by 71

The coefficient gender variable is negative and significantat 5 level of significance This result supports the appro-priate expectation that female respondents have a higherlikelihood of willing to pay for improved waste managementservices as compared to their male counterparts This is par-ticularly so because in Ghana women are mainly responsiblefor waste management at the household level

53 The Amount of Money Respondents Are Willing to Payfor Improved Waste Management Services Households whowere paying were asked to indicate how much more they arewilling to pay and the result is presented in Figure 1

From Figure 1 it could be noted that close to 50 ofthe households who are currently paying for solid wastemanagement services are willing to pay GHC 500 more forthe hypothetical situation described for waste managementservices Very few (about 63) were willing to paymore thanGHC 1000

Thus the majority of the respondents are willing to con-tribute at least GHC500more in support of the improvementin the waste management services within the metropolis


Table 3 Tobit regression results of factors influencing the amount of money respondents are willing to pay for improved waste managementservices

Independent variable Coefficient Std errors 119875 gt 119911 Marginal effectAge 02376 00854lowastlowastlowast 0005 04724Gender 30367 22908 0185 08780Marital status minus06668 29889 0823 08780Number of children minus05199 06764 0442 20731Income minus00100 00018lowastlowastlowast 0000 01126Education 05215 02331lowastlowast 0025 146341Length of stay minus05231 01669lowastlowastlowast 0002 89268House ownership 61376 21265lowastlowastlowast 0004 07560Bags of waste generated 02996 00993lowastlowastlowast 0003 173171Distance minus00172 00059lowastlowastlowast 0004 1828540

Goodness of fit measures119877-squared 06962Adj 119877-squared 06714lowastlowastlowastSignificant at 1 lowastlowastsignificant at 5

Histogram

0

10

20

30

40

50

Freq

uenc

y

0 10 20 30 40 50Ghana Cedis

Mean = 813 std dev = 7884119873 = 140

Figure 1 The additional amount of money households are willingto pay

The mean or average amount of money the respondents arewilling to pay is GHC 813

The next section investigates the factors which influencethe amount of money the respondents are willing to pay forimproved waste management services

54 Determinants of the Amount of Money Households AreWilling to Pay for ImprovedWaste Management Services TheTobit regression results of factors influencing the amount ofmoney respondents are willing to pay for improved wastemanagement are presented in Table 3 To determine whichof the factors identified using the logit model influencesthe amount of money the respondent are willing to pay forimproved waste management services the truncated Tobitmodel was usedThe truncation was as a result of the fact thatthose respondents who are not willing to pay for improvedwaste management services were excluded from the ana-lysis The Tobit regression results gave an adjusted 119877-squaredof about 067which implies that at least one of the explanatoryvariable included in the model has coefficient differentfrom zero Giving this goodness of fit measure (adjusted

119877-squared) it can be concluded that the Tobit model usedin reliable and has the requisite explanatory power All theincluded explanatory variables met the apriori expectations

The coefficients of age variable shows positive and signif-icant relationship with the amount of money the respondentsare willing to pay for improved waste management and it issignificant at 5 This may be explained by the fact that aspeople get older they tend to understand the need of a cleanenvironment In addition they may also know that access tofunds by waste management organization can improve theirservices As age increases by one year the amount of moneyindividual would be willing to pay would increase by 47The coefficient of household income is positive and signi-ficant This result is in agreement with the environmentaleconomics literature on the positive relationship betweenincome and the demand for improvement in the environmen-tal quality [17]

A GHC 1 increase in household income is likely toincrease the amount of money the respondents are willingto pay by 1126 This result shows that there is a positiverelationship between education and the amount of moneythe respondents are willing to pay for improving waste man-agement services This may be explained by the opportunityeducation gives to people to understand the consequence ofimproper waste disposal

In Ghana cost of housing is high and moreover land-lords are persons held responsible for an unclean house incase the actual cause of the filth is not immediately identifiedThus it is not surprising that respondents living in their ownhouses would pay a higher amount of money for improvedwaste management services

The longer the respondents stay in a particular commu-nity the higher the amount of money they are willing topay for improved waste management services This couldbe because the longer stay in the area would help one tounderstand the problem of sanitation in the area better andwhip up the need for proper sanitation of the area in the indi-vidual Additional year stay in a house would result in892 increase in the amount of money the respondents are


willing to pay The volume of waste generated has a posi-tive and significant relationship with the amount of moneyrespondents are willing to pay This can be explained bythe fact that those who generate larger volume of wastewould have more problems with disposal and hence wouldbe willing to pay more for its disposal

Distance to solid waste dumping sites has the expectedsign and is significant at 5 As the distance to the dumpingsites increases the amount of money the respondents are will-ing to pay also increases This is because increase in distancecomplicates the problem of solid waste management as peo-ple would have to walk far distances to dispose of waste

6 Conclusion and Policy Implications

Respondents were willing to pay more for improved wastemanagement services The determinants of willingness topay for improved waste management services were identifiedusing logit regression model Level of education length ofstay in the area housing arrangement and distance to solidwaste dumping sites as well as gender were noted to signifi-cantly influence the respondentsrsquo likelihood of willingness topay for improved waste management services

Generally the majority of the respondents are willing topay GHC 500 more for improved waste management ser-vices Again the factors influencing the amount ofmoney res-pondents are willing to pay for improved waste managementservices were determined using the Tobit model The signif-icant factors include age income education length of stayhouse ownership bags of waste generated and distance todumping sites

The assembly should take advantage of the fact that resi-dents of the metropolis see waste management as a collectiveresponsibility and not the sole role of the government It musttherefore endeavour to provide tailor-made services and pos-sibly preinvest to provide first class services (improved wastemanagement services) as the public pledge their preparednessto pay when the services are improved

Income level of individuals was realized to determinewillingness to pay more for improved services The assemblycan therefore abolish the flat rate payment system andsurcharge the first and second class residential areas to payrelatively more (because they can afford) and use the excessamount of money to subsidize the third class residential areas(because they cannot afford)

The sanitation bylaws of the KMA need an urgent reviewas some sections seem to be outmoded (eg fine of GHC500 for sanitation offenders) Copies of the revised bylawsmust not be kept under lock and key as is currently beingdone but rathermade readily available to the public Abridgedversions should be printed and distributed to all citizens ofthe metropolis It must also be published on the assemblyrsquoswebsite for easy access by internet users

The assembly should invest in educating the public tho-roughly to understand the impact of waste on the socioecon-omic development of a nation and the roles of the individualsThey should also be made to understand well the polluterprinciple and why it is necessary and the contents of thebylaws and their implications

References

[1] U S Environmental Protection Agency ldquoInternational WasteActivitiesrdquo 2009 httpepagov

[2] United Nations Environment Programme ldquoWaste Quantitiesand Characteristicsrdquo pp 31ndash38 2005 httpwwwuneporjp

[3] KMA ldquoKMArdquo Kumasi Metropolitan 2006 httpwwwkmaghanadistrictsgovgh

[4] T Amemiya ldquoQualitative response model a surveyrdquo Journal ofEconomic Literature vol 19 pp 1483ndash1536 1981

[5] L Hill and P Kau ldquoAnalysis of purchasing decision with multi-variate probitrdquo American Journal of Agricultural Economics 53no 5 pp 882ndash883 1981

[6] R S Pindyck and D L Rubinfeld Econometric Models andEconomic Forecasts McGraw- Hill New York NY USA 2ndedition 1981

[7] W H Green Econometric Analysis Prentice Hall Upper SaddleRiver NJ USA 4th edition 2008

[8] M A Akudugu I S Egyir and A Mensah-Bonsu ldquoWomenfarmersrsquo access to credit from rural banks in Ghanardquo Agricul-tural Finance Review vol 69 no 3 pp 284ndash299 2009

[9] M Ayamga D B Sarpong and S Asuming-Brempong ldquoFac-tors influencing the decision to participate in microcredit pro-grammes an illustration for Northern Ghanardquo Ghana Journalof Development Studies vol 3 no 2 pp 57ndash65 2006

[10] O I Oladele ldquoA Tobit analysis of propensity to discontinueadoption of agricultural technology among farmers in South-western Nigeriardquo Journal of Central European Agriculture vol6 no 3 pp 249ndash254 2005

[11] A A Dankyi P Y K Sallah A Adu-Appiah and A Gyamera-Antwi ldquoDetermination of the adoption of quality proteinmaizeObatanpa in Southern Ghanamdashlogistic regression analysisrdquo inProceedings of the 5th West and Central Africa Regional MaizeWorkshop IITA Cotonou Benin May 2005

[12] M A Akudugu I S Egyir and A Mensah-Bonsu ldquoAccess torural bank in Ghana the case of women farmers in the UpperEast Regionrdquo Ghana Journal of Development Studies vol 6 no2 pp 142ndash167 2009

[13] C A Cameron and P K Trivedi Microeconometrics Methodsand Applications Cambridge University Press New York NYUSA 2005

[14] J Baidu-Forson ldquoFactors influencing adoption of land-enhanc-ing technology in the Sahel lessons from a case study in NigerrdquoUnited Nations University Institute of Natural Resources inAfrica University of Ghana Legon Ghana 1999

[15] O Zerbock ldquoUrban SolidWasteManagementWaste Reductionin Developing Nationsrdquo 2003 httpwwwceemtuedu

[16] C Robson Real-World Research A Resource for Social Scientistsand PractitionermdashResearchers Blackwell Malden Mass USA1993

[17] L Medina Some Myths about the Collective Action ProblemDepartment of Political Science University of Chicago 2002

Submit your manuscripts athttpwwwhindawicom

Child Development Research

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Education Research International


Hindawi Publishing Corporationhttpwwwhindawicom

Volume 2014

Biomedical EducationJournal of

Psychiatry JournalHindawi Publishing Corporationhttpwwwhindawicom Volume 2014

ArchaeologyJournal of


AnthropologyJournal of



Research and TreatmentSchizophrenia


Urban Studies Research

Population ResearchInternational Journal of


CriminologyJournal of


Aging ResearchJournal of



NursingResearch and Practice

Current Gerontologyamp Geriatrics Research


Volume 2014

Sleep DisordersHindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Journal of AddictionHindawi Publishing Corporationhttpwwwhindawicom Volume 2014


Volume 2014

Economics Research International

Depression Research and TreatmentHindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Geography JournalHindawi Publishing Corporationhttpwwwhindawicom Volume 2014


Research and TreatmentAutism


also subscribe to specialty services like medical waste pick-up services or confidential paper shredding and disposalservices

Waste management practices differ for developed anddeveloping nations for urban and rural areas and for resi-dential and industrial producers For instance in some casesmanagement for nonhazardous residential and institutionalwaste in metropolitan areas is usually the responsibility oflocal government authorities while management for hazard-ous commercial and industrial waste is usually the respon-sibility of the generator Developing effective waste manage-ment strategies is critical for nations all over the world asmany forms of waste can develop into a major problem whenthey are not handled properly Numerous firms provide wastemanagement services of a variety of types and several gov-ernments also regulate the waste management industry forsafety and efficacy

According to [1] historically the amount of waste gener-ated by human population in the early ages was insignificantmainly due to the low population densities coupled with thefact that there was very little exploitation of natural resourcesCommon wastes produced during the early ages were mainlyashes and human and biodegradable wastes and these werereleased back into the ground locally with minimal adverseenvironmental impact

Before the widespread use of metals wood was widelyused for most applications However reuse of wood has beenwell documented Nevertheless reuse and recovery of suchmetals have been carried out by earlier humans With theadvent of industrial revolution waste management became acritical issue This was due to the increase in population andthemassive migration of people to industrial towns and citiesfrom rural areas There was a consequent increase in indus-trial and domestic wastes posing threat to human health andenvironment

In Africa municipal solid waste management constitutesone of the most crucial health and environmental problemsfacing governments of African cities This is because eventhough these cities are using 20ndash50 percent of their budgetin solid waste management only 20ndash80 percent of the wasteis collected The uncollected or illegally dumped wastes con-stitute a disaster for human health and the environmentaldegradation Not only is their quantities increasing but theirvariety is also increasing both a consequence of increasingurbanization incomes and changing consumption habitsfuelled by globalizationThis scenario places the already des-perate urban councils in a difficult situation especially as theyhave to develop new strategies to deal with increasing vol-umes as well as strange varieties of wastes Poor waste man-agement practices in particular widespread dumping ofwaste in water bodies and uncontrolled dump sites aggra-vates the problems of generally low sanitation levels acrossthe African continent

According to [2] urbanization is on the rise inAfrica andthis trend is expected to continue in the future Of concern isthe inability of infrastructure and land use planningmethods(including waste management) to cope with urban growth(the highest in the world) at 35 percent annuallyThis is part-icularly urgent in slum areas which constitute a big part of

many of the cities and towns in Africa Waste managementinfrastructure is largely nonexistent in rural areas of Africa

Imports of second-hand consumer goods and productionandor import of substandard products are all contributing tothe rapid increase in waste generation Policies should be putin place and existing standards enforced to reverse this trendImplementation or enforcement of waste regulations andconventions is severely constrained by the lack of good gov-ernance transparency and prevalence of corruption in somecases Lack of awareness and appreciation of best practicesfor environmentally sound management of wastes is a majorconstraint A paradigm shift among communities and societyat large is needed Again the fast growing use of ICT andrapid turnover in technology (particularly computers mobilephones etc) is creating a growing E-waste stream for whichthere is no waste management capacity yet This leads to dis-posal of both E-waste and municipal waste in dump sitesChanging lifestyles and consumption patterns of the growingurban middle class in particular increases the complexityand composition of waste streams in Africa

The gap between waste management policy and legisla-tion and actual waste management practices is widening dueto perennial capacity constraints and lack of waste manage-ment facilities for various waste streams Access to majorinvestments and acquiring the technical know-how neededto resolve the capacity constraints remain a tall order Wastegeneration is expected to increase significantly as a result ofindustrialization urbanization and modernization of agri-culture in Africa This will further aggravate the current cap-acity constraints in waste management

Progress has been made in waste management policiesand strategies Biogas and compost production from organicwaste fractionation has been widely accepted in Africa as abest practice and progress is being made in developing andimplementing specific projects in various countries How-ever the use of economic instruments and implementationof polluter-pays principles in waste management have yet tomature in most African countries

The single largest implementation challenge formanagingwaste policies remains creating sufficient capacity for envi-ronmentally sound management including where appro-priate recovery and recycling of various waste streams acrossAfrica The effort to do this is constrained by access tofinance and technical know-how Current bylaws in mostAfrican countries give responsibility for waste managementto municipalities which are often ill-equipped to deal withcollection and disposal Such bylaws are now an impedimentto investment in waste management by the private sector

2 Problem Statement

Ghana in general andKumasiMetropolitanAssembly (KMA)in particular have made several attempts at addressing thewaste menace which is on the rise as a result of populationhikes growth in industrialization and consumer attitudesThe cityrsquos bylaws and policies on waste and sanitation seekto address the waste challenge in its entirety on individualbasis The assembly among its efforts and strategies has con-tracted some waste companies to handle waste collection and




3 Methodology





119884119894= 120573119883119894+ 120583119894 (1)



119884119894= 1 if119883


119883lowast and

119884119894= 0 if119883



Prob (119884119894= 1) = 119865 (120573

1015840119883119894)

Prob (119884119894= 0) = 1 minus 119865 (120573

1015840119883119894)

(2)







119875 (119884 = 1) =1198901205731015840119883

1 + 1198901015840119883

119875 (119884 = 0) = 1 minus1198901205731015840119883

1 + 1198901205731015840119883=

1

1198901205731015840119883

(3)


119864(119884

119883) = 1 [119865 (120573

1015840119883)] + 0 [1 minus 119865 (120573

1015840119883)] = 119865 (120573

1015840119883) (4)



120597119875119894

120597119883119894

= [1205821205731015840119883

1 + 1205821205731015840

119883]120573 = 119865 (120573

1015840119883) [1 minus 119865 (120573

1015840119883)] 120573 (5)




119910119894= 1198831015840120573 + 120576119894 (6)





119910119894= 0 if 119910lowast




agement services





119864119910= 119883119894120573119865 (119911) + 120590119891 (119911) (7)





119864119910= 119865 (119911) 119864

lowast

119910 (8)


120575119864119910

120575119883119894

= 119865 (119911) (

120575119864lowast

119910

120575119883119894

) + 119864lowast

119910(119865 (119911)

120575119883119894

) (9)




120575119864 (119910119883119894)

120575119883119894



ln 119871 = sum

119910119894gt0

minus1

2

[

[

log (2120587) + ln1205902 +(119910119894minus 1198831015840

119894120573)2

1205902]

]

+ sum

119910119894=0

ln[1 minus 120601 (119883

1015840

119894120573)

120590]

(11)

















Years

Housingarrangement

Housingarrangement


Number ofdependents


Number individuals









































Histogram

0

10

20

30

40

50

Freq

uenc

y

0 10 20 30 40 50Ghana Cedis

Mean = 813 std dev = 7884119873 = 140




















References
























Volume 2014





















Volume 2014




Volume 2014









3 Methodology





119884119894= 120573119883119894+ 120583119894 (1)



119884119894= 1 if119883


119883lowast and

119884119894= 0 if119883



Prob (119884119894= 1) = 119865 (120573

1015840119883119894)

Prob (119884119894= 0) = 1 minus 119865 (120573

1015840119883119894)

(2)







119875 (119884 = 1) =1198901205731015840119883

1 + 1198901015840119883

119875 (119884 = 0) = 1 minus1198901205731015840119883

1 + 1198901205731015840119883=

1

1198901205731015840119883

(3)


119864(119884

119883) = 1 [119865 (120573

1015840119883)] + 0 [1 minus 119865 (120573

1015840119883)] = 119865 (120573

1015840119883) (4)



120597119875119894

120597119883119894

= [1205821205731015840119883

1 + 1205821205731015840

119883]120573 = 119865 (120573

1015840119883) [1 minus 119865 (120573

1015840119883)] 120573 (5)




119910119894= 1198831015840120573 + 120576119894 (6)





119910119894= 0 if 119910lowast




agement services





119864119910= 119883119894120573119865 (119911) + 120590119891 (119911) (7)





119864119910= 119865 (119911) 119864

lowast

119910 (8)


120575119864119910

120575119883119894

= 119865 (119911) (

120575119864lowast

119910

120575119883119894

) + 119864lowast

119910(119865 (119911)

120575119883119894

) (9)




120575119864 (119910119883119894)

120575119883119894



ln 119871 = sum

119910119894gt0

minus1

2

[

[

log (2120587) + ln1205902 +(119910119894minus 1198831015840

119894120573)2

1205902]

]

+ sum

119910119894=0

ln[1 minus 120601 (119883

1015840

119894120573)

120590]

(11)

















Years

Housingarrangement

Housingarrangement


Number ofdependents


Number individuals









































Histogram

0

10

20

30

40

50

Freq

uenc

y

0 10 20 30 40 50Ghana Cedis

Mean = 813 std dev = 7884119873 = 140




















References
























Volume 2014





















Volume 2014




Volume 2014









119910119894= 1198831015840120573 + 120576119894 (6)





119910119894= 0 if 119910lowast




agement services





119864119910= 119883119894120573119865 (119911) + 120590119891 (119911) (7)





119864119910= 119865 (119911) 119864

lowast

119910 (8)


120575119864119910

120575119883119894

= 119865 (119911) (

120575119864lowast

119910

120575119883119894

) + 119864lowast

119910(119865 (119911)

120575119883119894

) (9)




120575119864 (119910119883119894)

120575119883119894



ln 119871 = sum

119910119894gt0

minus1

2

[

[

log (2120587) + ln1205902 +(119910119894minus 1198831015840

119894120573)2

1205902]

]

+ sum

119910119894=0

ln[1 minus 120601 (119883

1015840

119894120573)

120590]

(11)

















Years

Housingarrangement

Housingarrangement


Number ofdependents


Number individuals









































Histogram

0

10

20

30

40

50

Freq

uenc

y

0 10 20 30 40 50Ghana Cedis

Mean = 813 std dev = 7884119873 = 140




















References
























Volume 2014





















Volume 2014




Volume 2014


















Years

Housingarrangement

Housingarrangement


Number ofdependents


Number individuals









































Histogram

0

10

20

30

40

50

Freq

uenc

y

0 10 20 30 40 50Ghana Cedis

Mean = 813 std dev = 7884119873 = 140




















References
























Volume 2014





















Volume 2014




Volume 2014

























Histogram

0

10

20

30

40

50

Freq

uenc

y

0 10 20 30 40 50Ghana Cedis

Mean = 813 std dev = 7884119873 = 140




















References
























Volume 2014





















Volume 2014




Volume 2014
















References
























Volume 2014





















Volume 2014




Volume 2014












Volume 2014





















Volume 2014




Volume 2014






urban households' willingness to pay for improved solid waste

Documents