UNIVERSITY OF IBADAN LIBRARY
Jour, of Trap. /■'or. Krsnm ces V'ol. 1 9(2): 2003: 1 - 12
THE USE OFTOBIT MODEL IN EVALUATING PROSPECTIVE 
PARTICIPATORY FINANCING OF REHABILITATION OF IBADAN URBAN
FOREST RESERVES
AJEWOLE Opeyemi Isaac
Department of Forestry, Wildlife and Fisheries, Faculty of Agricultural Sciences 
University of Ado-Ekiti, Ado-Ekiti, Nigeria 
E-mail: openjc>vole@iustice.com, oncaicwole@vahoo.co.iik
ABSTRACT
This study evaluated the factors that affect public willingness 
to participate financially in the rehabilitation o f Ibadan urban 
forest reserves. A one limit Tobit model was used to analyse 
data obtained ifrom a multi-stage randomly sampled 397 
resfdents o f Ibadan metropolis. The results show that 90% of 
the respondents were willing to contribute funds to the project. 
Furthermore, the model reveals that a change in the independent 
variables would lead to a greater change (77%) in the 
probability of contributing money, relative to change (23%) in 
the amount of money to be contributed. Willingness to 
participate financially in rehabilitation of Ibadan urban forest 
reserves was found to be positively influenced by the variables: 
being employed,(p < 0.01). having trees in the residence
(p < 0.01), and being a past beneficiary o f the forests (p<0.05).
'S he results confirm the need to make the urban forest more 
directly beneficial to the public in order to harness the funds 
which the public is willing to contribute to its rehabilitation.
INTRODUCTION
Ibadan metropolis, comprising five local government areas namely: Ibadan North, Ibadan 
North-West, Ibadan North-East, Ibadan South-West and Ibadan South-East, used to have 
four forest reserves viz;. Oke-Aremo, Ogunpa Dam, Eleyele and Alalubosa forest 
reserves. They were constituted over a forty-year period (1916 to 1956) to provide 
fuelvvood and building poles for the metropolitan population. They were also to provide 
watershed protection for Ogunpa dam, Alalubosa Lake and Eleyele dam (Popoola and 
Ajewole; 2001, 2002). Currently, Alalubosa and Oke-Aremo reserves have been 
dereserved given way for various infrastructural developments in the metropolis, while 
Eleyele is a patch of highly degraded forest. Ogunpa dam Forest Reserve suffered similar 
fate, leaving only a small preserved area; the Agodi gardens which serves as recreation 
forest cum zoological garden.
;
UNIVERSITY OF IBADAN
LIBRARY
A.IEIVOI.I'.. ( )/>c\vnii Isom Jour. o f Trap. Tor. Resources I'nl. 19(2) 200} 1- 12
The government o f Oyo Slate recently made some efforts to rehabilitate the Agodi 
gardens in order to resuscitate the recreation functions o f the forest. However, sustained 
government funding of conservation programmes has often m et with set-backs. The 
reasons for this include competing demands for government funds and the little 
immediate direct moncy-benelils o f conservation projects. These obviously constrain the 
How o f funds to conservation projects, and therefore call for. the development of 
innovative funding mechanisms. Popoola (2004) observed that funding o f reforestation 
projects is last becoming a low priority issue of governments in most states o f Nigeria 
bringing about ridiculously low budgetary allocations which arc never released on time. 
Participatory funding o f conservation which is the focus o f this study can be a potent 
machinery for conservation projects, since it engenders a sense o f ownership in the minds 
o f the members o f (he public. The objective o f this study therefore is to determine the 
factors that affect public willingness to contribute financially and the prospective 
contributory amount for the rehabilitation o f II adan urban forest reserves.
DATA COLLECTION AND ANALYSIS 
Data and Sampling Procedure
The data for this study were obtained from a questionnaire survey o f 400 residents of 
Ibadan ’metropolis, using the payment card contingent valuation method. The 
questionnaire which was designed in line with lloelin (1902) and I lanemann (1994) 
elicited the respondents' willingness to contribute financially to rehabilitation o f the 
deg raded  forests of the metropolis. Data on the respondents' socio-economic 
characteristics were also collected. In all, 397 were successfully retrieved and used for the 
analysis. Multi-stage random sampling procedure was adopted in collecting the data. The 
metropolis was stratified into two: (i) neighbourhood; comprising the residents who live 
in areas within I km radius o f each urban forest reserve, and (ii) non-neighbourhood; 
comprising residents living outside I-km radius of each forest reserve. The 
neighbourhood stratum was further stratified into four cells, where the non­
neighbourhood stratum was strati lied to live. Names ol major streets in each ol the 
resulting nine cells were collected from the Metropolitan Planning Authority. Using a 
table o f random numbers, two streets were picked front each ol the nine cells.
Twenty live respondents were sampled in each ol the two selected streets ol lour cells in 
the neighbourhood stratum, while 20 respondents were drawn from each ot the tw'o 
selected streets of the live cells in the non-neighbourhood stratum. T his gave a sample 
size o f 400 respondents. Due to incomplete information in 3 of the questionnaire, they 
were discarded, flic effective sample si/.e used in the subsequent analysis was 397.
The Analytical Mode!
Participatory financing is conceptualised to involve a two step simultaneous decision by 
respondents. This entails:
i. ' whether or not to participate in financing the scheme; and
ii the amount of financial contribution the decision maker is willing to make alter the 
initial decision has been made.
UNIVERSITY OF IBADAN LIBRARY
AJEWOj.E. Opcyemi Isaac, low. of Trap. Ear. Resources IW. 19(2): 2003: t - 12
The Tobil model (a hybrid of probit and multiple regression models) is appropriate in 
capturing such a decision process (Tobin; 1958).
Model Specification
The one-limit Tobit model for this study is specified as follows:
c = c *  = X,p + Ci ifXiP + e ,>  0
= 0 if Xip + ej = < 0 .................... equation (1)
Where;
i 1,2... n observations.
X, the vector of explanatory variables.
C = limited dependent variable
C* = continuous dependent variable which is observed only when it is 
positive.
P = vector of unknown coefficients
e, = error term ;~N 1(0,a 2)
McDonald and Moflit (1980) argued that the Tobil coefficients do not correctly measure the 
regression coefficients for observations above the limit. Therefore the Tobit coefficients 
have to be decomposed in order to isolate the probability of participating financially and the 
amount which the prospective participant will like to contribute to rehabilitation of Ibadan 
urban forest reserves, 'flic procedure for the decomposition according to McDonald and 
MolTit (1980) is as follows;
The expected value of C in the model is;
EC-X\\\\z)  +af(z)....................................................  equation (2)
Where z = Xp/o,
f(z) = the unit norma! density,
F(z) is the cumulative normal distribution function, 
a = standard error of estimate of the dependent variable C.
X and p are as previously defined in equation (1).
Furthermore, the expected valpe of C for those willing to participate in contributing funds to 
rehabilitation of the degraded Ibadan urban forest reserves (i.e. prospective contributory 
amount) is given as
£C*= Xp+ of(z)/F(z)........................................................... equation (3)
Consequently, the basic relationship between the expected value of all observations; EC, the 
expected value conditional on willingness to contribute funds to rehabilitation of the 
degraded Ibadan urban forest reserves; EC'' (i.e. the expected contributory fund), and the 
probability of contributing; F(z) is
EC = F(z) EC* .........................................................  equation (4)
If we consider the effect of a change in the ilh explanatory variable ‘X’ on C:; the 
disaggregation of C is given as:
UNIVERSITY OF IBADAN LIBRARY
A.JI'AVOI.E, Opeyemi'lxaac/Jour, o f Trap. For. Resources Vol. 19(2): 2003: 1- 12
. SEC / SX, = F{z)(5EC * / SX,) + EC * (SF(z) / SX, ) ............. . ......  equation (5)
Equation (5) shows the total change in C being disaggregated into two parts: (1) the change 
in C of those that are willing to contribute money (expected value of amount to be 
contributed weighted by the probability of contributing money), and (2) the change in the 
probability of contributing money weighted by the expected value of the amount of money 
to be contributed.
As the independent variable xiciwngc in the probability of contributing funds to rehabilitation 
of Ibadan urban forest reserves changes is;
SF{z)/SXl = f{z)P, I cr................................................ equation (6)
And the change in contributory amount with respect to a change in an explanatory variable 
among the participants is
SE(C') / SX, = - z j \ z ) l  F ( z ) - f ( z ) 1 / F(z)2}  ...............  equation (7)
(3 and o were obtained from the maximum likelihood estimates (MLE) and used to compute 
each of the terms in equations (5) (6), and (7). LIMDEI’ (version 7) was employed in 
running the analysis for the maximum likelihood estimates (MLE) o f the Tobit coefficients, 
while SAS (version 8) was used to write the program for the decomposition of the MLE.
The Empirical Tobit Model
In evaluating the determinants of participatory financing o f rehabilitation of Ibadan urban 
forest reserves, an empirical censored Tobit model comprising 24 socio-economic 
regressors was specified. These include income (which is made up o f four dummy 
variables; no income, low income, medium income, and high income), age (which 
consists o f five dummy groups; 15-24, 25^34, 35-44, 45-54, 55 and above), education 
(also comprising four dummy variables; illiterate, primary education, secondary 
education and tertiary education), employment, gender, marriage, proximity of the 
residence to the urban forest reserves, ethnicity, duration of residence in the study area, 
awareness of the existence of the urban forest reserves, being a past beneficiary of the 
urban forest, having interest in planting trees, having trees in one’s residence and having 
interest in joining organisation that will mobilise for the rehabilitation project. Each of the 
variables; income, age and education was divided into different classes in order to capture 
the possible effect of each of the classes within each of the partitioned variables. For 
instance, different income groups have varying financial empowerment, so also do 
different age groups have varying interests and financial empowerment. Similarly people 
with different levels of education do often have different perspectives or dispositions to 
certain issues. With the exception of the variable -  duration of residence in the study 
area- all other variables are dummy.
UNIVERSITY OF IBADAN LIBRARY
AJl'M'Ol. Opeyemi IsaacJour, o f Trap. For. Resources I'ol 19(2): 2003: / - 12
RESULTS AND DISCUSSION
'Hie results of the empirical models are presented in Tables I and 2. The Maximum 
Likelihood Estimate (M I.li) coefficients of the I obit regression are as shown in Table I 
Irom which can be observed that only 21 regressors, excluding the constant, were used to 
run the I obit analysis. I he remaining three variables were used as control. Thus illiteracy 
was used as control lor the education variables, no income for the income variable and age 
class 1 (age range 15-24) was used as control for the age variables furthermore, the results 
in Table i also show that 7 regressors were found to have significant influence on public 
willingness to linancc the rehabilitation of the degraded urban forest reserves. These are 
age class 2; comprising 25-34 years group, (p<0.l), employment (p<0.1), medium 
income (p*-0.05), marriage (p < 0.1), being a past beneficiary of the urban forest reserves 
(p < 0.05), having trees in the respondent's residence (p < 0.05) and having interest in 
joining the mobilising organisation (p < 0 .1).
Using only the significant regressors to run the empirical model gave the result in Table 
2. The results indicate that age class 2 which was initially significant (p < 0.1) in Table I 
was no loger significant in this new model. It is also observed that the level of 
significance of some of the variables has increased. For example, the significance of 
“employment” increased tremendously from p < 0.1 to p < 0.01, while “having trees in 
the residence” and “ interest in joining the mobilising organisation” had their levels of 
significance increased from p < 0.05 and p < 0 .1 to p < 0.01 and p < 0.05 respectively.
UNIVERSITY OF IBADAN LIBRARY
A J  E l VOLE, O peyem i Isaac',lour, o f  Trap. Tor. R esources  [ '<)/. 19(2): 2003: l  - 1 2
Tabic I: Maximum Likelihood Estimate (IMLIC) Coefficients of the Tobit Regression
Variable Coefficient Standard lirrnr Normalised I’ll / - M
Coefficient
Constant 12.2376 37.4053 .327 .7435
g i:n d i ;r X , -.1195 .1095 -1.092 .2749
PROXIMITY -.0007 .0642 -.010 .9919
Atilt . -30.8837* 18.3253 -1.685 .0919
Cl.ASS-2 X.,
Atilt 20.0260 21.95 17 .912 .3616
Cl.ASS-3 x„
Atilt 39.8836 29.7034 1.343 .1794
Cl.ASS-4 x .
Atilt -28.9399 36.2782 -.798 .4250
Cl.ASS-5 x 6
ItMI’l .OYMItNT X 7 40.3546* 23.3851 1.726 .0844
II it ill INCOMli X 8 40.2607 30.7903 1.308 .1910
MltDIUM INCOMli Xy -60.6047+* 26.3183 -2.303 .0213
LOW INCOMli X|o 19,9859 17.9421 1.114 .2653
MARRIAGli X , , -.3446* .1987 -1.734 .0830
PRIMARY liOUCA tion 20.3956 30.6962 .664 .5064
X|2
SliCONDARY -17.1378 18.6970 -.917 .3593
liOl/CATION X n
TliRTlARY -3.2060 18.1986 -.176 .8602
lilHJCATION X,.,
li 11 INK’ X u -.1292 .1315 -.982 .3259
DURATION X !f) -.0396 .07693 -.5 15 .6069
AWARIiNliSS X|7 13.6434 28.4329 .480 .63 1 3
PAST BliNliFICIARY X|« 61.7564+* 30.9161 1.998 .0458
1NT liRIiS I IN TRIi ii -.0447 .1065 -.420 .6746
PLANTING X  |9
I RItr.S IN RliSIDIiNCIi 41.7955* + • 18.9607 2.204 .0275
X20
INUiRliS I IN -.3300+ .1774 -1.860 .0629
ORGANISATION X 2|
Siomn 175,4192+ + + 7.6130 23.042 .0000
* :  significant at 10%, * * : significant at 5%, * * * :  significant at 1%
6
UNIVER XSITY OF IBADAN LIBRARY
AJliWOLH, Opeyemi Isaac Jour, o f  Trap. T'or. Resources To/. 19(2): 2003: t - 12
Table 2: McDonald and Moffit Decomposition of the Empirical Tobit Model
for Prospective Participatory Financing of Rehabilitation of the 
Ibadan Urban Forest Reserves.
V ariab le C u e  r t i c ie n t N o rm a l is e d P - M e a n  o f  X Total ■ C h a n g e  in  C h a n g e  
C o e f f ic ie n t V a lu e c h a n g e th e in  th e  
SEC /  c fV , c o n t r ib u to ry p ro b a b i l i ty  
a m o u n t o f
SEC' ISX, c o n t r ib u t in g
f in a n c e
SFizj /SY,
C o n s ta n t -3 .8 9 3 3 - .1 2 8 .8 9 8  1 - 0 .1 6 3 9 -0 .5 1 3 8 - 0 .0 0 1 9
K M P i.O Y M h N T 5 1 .0 0 9 9 * + * 2 .6 2 6 .0 0 8 6 -1 .8 4 9 3 2 .1 4 8 6 6 .7 3 7 7 0 .0 2 5 5
x ?
M I-.D H IM  X y - .3 8 9 0 * * -2 .1 5 8 .0 3 1 0 -4 .9 2 2 1 -0 .0 1 6 4 -0 .0 5 1 5 - 0 .0 0 0 2
IN C O M I-:
M A R K IA d l  '. X 11 - .3 3 1 1  + -1 .8 5 7 .0 6 3 3 -4 .5 6 U 3  . -0 .0 1 3 9 - 0 .0 4 3 6 - 0 .0 0 1 7
PA ST ' X f x 6 7 .2 5 1 3 * * 2 .2 0 9 .0 2 7 2 - 1 .5 9 8 0 2 .8 3 2 7 8 ,8 8 2 8 0 .0 3 3 7
1 H K N IT IC 1 A R Y
m i d i s  IN  X 2() 5 2 .4 5 2 8 * * * 2 .8 5 3 .0 0 4 3 -2.0201 2 .2 0 9 3 6 .9 2 7 9 0 .0 2 6 3
KHSIDFNCF
IN T K R K S T IN  X 2f - .3 7 8 0 * * -2 .0 9 8 .0 3 5 9 -4 .2 6 3 8 0 .0 1 6 0 0 .0 5 0 2 0 .0 0 0 2
O R G A N I S A T I O N
■Sigma 1 7 9 .4 5 * * * 7 .7 8 6 1 2 3 .0 5 .0000
*: significant at 10%. **: significant at 5%, ***: significant at 1% 
z = -1.73, f(z) = 0.04, F(z) = 0.90
Table 2 also presents McDonald and Moffit decomposition of the significant regressors 
which was carried out to evaluate the change in the probability o f contributing money as 
well as the change in the amount o f money respondents are willing to contribute to the 
rehabilitation o f the degraded Ibadan urban forest reserves, that may be occasioned by the 
change in any of the significant regressors.
Following McDonald and Moffit (1980), the value o f F(z) which equals 0.90, implies that 
90% of the observations have non-zero willingness to contribute money to rehabilitation of 
the degraded Ibadan urban forest reserves, in other words, 90% of the respondents are 
willing to contribute money to rehabilitation of the degraded Ibadan urban forest reserves, in 
order lo'evaluale the prospective participatory financing of the degraded Ibadan urban forest 
reserves at this point requires finding the value of equation 8 below;
u=.[ 1 -zf(z)/F(z)-f(z)2/F(z)2|....................................................................  equation 8
Putting in the required parameters; 
a  = 0.23
UNIVERSITY OF IBADAN LIBRARY
AJEWOI.K, Opeyemi Isaac/Jow. of Trop. For. Resources I'ol. 19(2): 2003: I - 12
The implication of Ihe above is that 23% of (he change in willingness to contribute money to 
the rehabilitation of the degraded Ibadan metropolitan forest reserves owing to change in the 
independent variables will be generated by change in the amount of money to be contributed 
while 77% will be generated by a change in the probability of contributing money, In 
essence, this implies that a change in independent variables will lead to greater change 
(77%) in the probability of contributing money, relative to a change (23%) in the amount of 
money to be contributed.
From Table 1, it can be observed that only 2! independent variables were used to run the 
model. The remaining three variables were used as control, viz: the variable illiteracy was 
used as control for the educational variables, no income was used as control for “income’' 
and lastly the variable 15-24 years as control lor age, since all these are categorical 
variables. This procedure is consistent with Gujarati (1985). Table 2 shows an intercept 
value of -0.1639 which is the autonomous total change in willingness to contribute money to 
tire rehabilitation of Ibadan degraded urban forests; decomposed to -0.5138, the autonomous 
change in the amount of money respondents are willing to contribute to the project and 
0.0019, the autonomous change in the probability of contributing money to the project.
fhe results further show that the variable; “employment" significantly (p < 0.01) and 
positively affects both the probability of contributing as well as the amount of money 
respondents are willing to contribute to the rehabilitation project, 'fhe coefficient of the 
intercept dummy of the respondents who are employed and are willing to contribute 
finances to rehabilitation of the degraded urban forest reserves is 6.7377. This implies that 
relative to those who are not employed; the contributory amount to the rehabilitation of the 
degraded Ibadan urban forest reserves is expected to increase by N6.73 for the employed 
than those who are not employed. Similarly, the coefficient of the intercept dummy for those 
who'are employed but are not willing to contribute finances to the rehabilitation project is 
0.0255. This also implies that relative to those who are not employed; the level of 
autonomous probability (0.0019) of contributing finances will increase by 0.0255, hence 
having an autonomous probability of contributing finances of 0.0274, as against those who 
are not employed. Therefore, the probability of an employed person to participate in 
contributing finances to rehabilitation of the degraded Ibadan urban forest reserves is 2.7%, 
while that of an unemployed person is 0.19%.
Out of the three variables that represented income, it is only the variable- medium income 
that .has significant (p<0.05) though negative influence on the willingness to contribute 
finances for the rehabilitation of Ibadan urban forest reserves. The coefficient of the 
intercept dummy of the respondents who are in the medium income category and are willing 
to contribute finances to rehabilitation of the degraded urban forest reserves is -0.0515.1 his 
implies that relative to those who are not earning income; the contributory amount to the 
rehabilitation of the degraded Ibadan urban forest reserves is expected to decrease by NO.05 
’for those in medium income category than those who are not earning income. Similarly, the 
coefficient of the intercept dummy for those in the medium income category that are not 
willing to contribute finances to the rehabilitation project is -0.0002. This also implies that
H
UNIVERSITY OF IBADAN LIBRARY
AJT'.WQIA:, O p eyem i Isaac, Jour, o f  Trap. T'or. R e so u rces  To/. 19(2): 2003: 1 - 12
relative to those who.are not earning income; the level of autonomous probability (0.0019) 
ol contributing finances will decrease by 0.0002, hence having an autonomous probability of 
contributing finances of 0.0017, as against those who are not earning income. The 
implication ol these is that the probability of a person in the medium income category to 
participate in contributing finances to rehabilitation of the degraded Ibadan urban forest 
reserves is 0.! 7%, while that of a person who is not earning income is 0.19%.
The variable “marriage” also has significant (p<0.1) but negative influence on willingness to 
contribute finances to rehabilitation of the degraded Ibadan urban forest reserves. The 
coefficient of the intercept dummy of the respondents who are married and are willing to 
contribute finances to rehabilitation of the degraded urban forest reserves is -0.0436. This 
implies that relative to those who are not married the contributory amount to the 
rehabilitation of the degraded Ibadan urban forest reserves is expected to decrease by 0.04 
Naira, as against those who are not married. Similarly, the coefficient of the intercept 
dummy for those who are married but not willing to contribute finances to the rehabilitation 
project is -0.0017. This also implies that relative to those who are not married; the level of 
autonomous probability (0.0019) of contributing finances will decrease by 0.0017, hence 
having an autonomous probability of contributing finances of 0.0002, as against those who 
are not married. The implication of these is that the probability of a married person to 
participate in contributing finances to rehabilitation of the degraded Ibadan urban forest 
reserves is 0.02% while that of an unmarried person is 0.19%.
Furthermore, the variable; “past beneficiary" has significant (p < 0.05) and positive 
influence on willingness to contribute finances to rehabilitation of the degraded Ibadan 
urban forest reserves. The coefficient of the intercept dummy for those who have previously 
enjoyed some benefits from these reserves and are willing to contribute finances to their 
rehabilitation is 8.8823. This implies that relative to those that have not previously enjoyed 
any benefit from these reserves, the contributory amount to the rehabilitation of the degraded 
Ibadan urban forest reserves is expected to increase by N8.89. Similarly, the coefficient of 
the intercept dummy for those who have previously enjoyed some benefits from the urban 
forest reserves, but not willing to contribute finances to their rehabilitation is 0.0337. This 
also implies that relative to those who have not previously enjoyed any benefit from the 
forest reserves, the level of autonomous probability (0.0019) of contributing finances will 
increase by 0.0337, hence having an autonomous probability of contributing finances of 
0.0356. Therefore the probability of a person who has enjoyed some benefits from Ibadan 
urban forest reserves is 3.56%, while that of a person who has not previously enjoyed any 
benefit from these reserves, to contribute finances to their rehabilitation is 0.19%.
Having trees in residence as a variable also has highly significant (p <0.01) and positive 
influence on the willingness to contribute finances to the rehabilitation of the degraded 
Ibadan urban forest reserves. The coefficient of the intercept dummy for the respondents 
whd have trees in their residences and are willing to contribute finances to the rehabilitation 
of the degraded fores! reserves is 6.9279. This implies that the contributory amount to the 
rehabilitation project is expected to increase by N6.93 as against those who do hot have trees
9
UNIVERSITY OF IBADAN LIBRARY
AJI'AVOLE, Opeyemi Isaac/Jow. of Trap. For. Resources i'nl. 19(2): 2003: I - 12
in their residences. Similarly, the coellicienl of the intercept dummy for those who have 
trees in their residences, but not willing to contribute to the rehabilitation of these degraded 
forest reserves is 0.0263. This implies that relative to those who have no trees in their 
residences, the level of autonomous probability (0.0019) of, contributing finances will 
increase by 0.0263, hence having an autonomous probability of contributing finances of 
0.0282. Therefore the probability of a person who has trees in his residence to contribute to 
the rehabilitation of Ibadan urban forest reserves is 2.82%.
The last variable in the empirical model- “interest in joining an organisation that will 
mobilise for the rehabilitation project-11 also has significant (p<0.05) but negative influence 
on willingness to contribute finances to the rehabilitation of the degraded Ibadan urban 
forest reserves. T he coellicienl of the intercept dummy for those who are interested in 
joining the organisation and are also willing to contribute to the rehabilitation project is - 
0.0502. This implies that relative to those who are not interested in joining the organisation, 
the.contributory amount to the rehabilitation project is expected to decrease by NO,05, as 
against those who are not interested in joining tire organisation. Similarly, the coefficient of 
the intercept dummy for those who arc interested in joining the organisation but not willing 
to contribute to the rehabilitation project is -0.0002. This also implies that relative to those 
who are not interested in joining the organisation, the level of autonomous probability 
(0.0019) of contributing finances to the project will decrease by 0.0002, hence having an 
autonomous probability of contributing finances of 0.0017. Therefore the probability of a 
person who is interested in joining a charitable organisation that will mobilise for the 
rehabilitation of the degraded Ibadan urban forest reserves is 0.17%, while that or the person 
who is not interested in joining such an organisation to contribute money to the 
rehabilitation project is 0 .19%.
CONCLUSION
This study has revealed a great potential for non-governmental funding of rehabilitation of 
the degraded Ibadan urban forest reserves. Though the modal contributory amount of N100 
per capita appears quite small, it is an amount that appears generally affordable across 
various socio-economic groups, and its aggregate value if properly harnessed can even be 
higher than the state's budgetary allocation for the development of the urban forest reserves. 
Be that as it may, harnessing these funds requires putting in place an effective fund 
collection mechanism. Some of these mechanisms include lotteries and sales of conservation 
souvenirs. Since being employed has significant positive influence on public willingness to 
participate financially in the rehabilitation of the degraded forests, the slate can organise 
lotteries and sales of souvenirs to target different categories of workers. Furthermore, 
participatory financing of the rehabilitation of Ibadan urban forest reserves should no doubt 
go side by side with appropriate participatory urban forestry development model which 
should ensure adequate power devolution and decentralisation that will guarantee optimum 
benefits for all the stakeholders. This model might be in form of a board or a 
jlleering/management committee for mobilisation, collection and management of funds for 
the implementation of the rehabilitation project. This-board that should comprise the
10
UNIVERSITY OF IBADAN LIBRARY
AJEWoiE, Opeyemi lstiac/Jour. of Trap. For. Resources Vol. /9(2): 2003: J - 12
representatives of various segments of the public can have its members drawn from the
academia, relevant professional groups, environmental and non-governmental organisations,
community based development organisations, the civil service, etc. This board will be
responsible for the day to day management of tile forest reserves and the funds. In a similar
manner, since a high proportion (76%) of the respondents are interested in joining a
charitable organisation that will mobilise for the funds, there is a need for a reputable 11011-
governmental organisation to take lip the challenge of mobilising or starting a parallel
organisation that will be responsible for mobilising of funds for participatory financing of
this rehabilitation project from the public.
One important factor that is germane to participatory financing is the need for appropriate
and adequate public enlightenment/education on the importance of the urban forest reserves
to the socio-economic well-being of the metropolitan residents. To this effect, government
needs to embark on result oriented public enlightenment programme about the need to
rehabilitate and conserve these urban forest reserves. More importantly, since previous use
of the urban forest has signi Iicant influence on willingness to participate financially in the
rehabilitation of the degraded urban forests, government should manage the urban forests in
a way to serve the diverse needs of the public. In this regards, recreation can be one of the
principal reasons Ior the rehabilitation of these forests, and the public should be encouraged
to make use of these forests for recreation and other needs. This will no doubt increase the
value and the importance of these forests in the minds of the public .
. I
REFERENCES
Gujarati, D.N (1985): "Basic Econometrics" McGraw-Hili Book Company lnc, 2nd
Edition. pp 431-504.
Hanemann, W.M. (1994): "Valuing the Environment through ContingentValuation".
The Journal of Econo IIIic Perspectives, 8(4), 19-43.
Hoehn, J.P. (1992): "Contingent Valuation in Fisheries Management; the Design of
Satisfactory Contingent Valuation Format". In Markandaya, A and Richardson,
.J. (eds.). The Earthscan Reader in Environmental Economics. pp 10) -) 11
(London Earthscan Publications Ltd.)
McDonald. J.F. and R. A. Mom! (1980): "The Uses or Tobit Analysis". The Review of
Economics and Statistic. Vol. LXII, 2, pp 318-321
Popoola . L. (2004): "Participatory Forest Finance in Nigeria: Problems and Prospects"
Paper Presented at the Workshop on Lessons Learnt on Sustainable Forest
Management in Africa, 09-13 February 2004, Nairobi, Kenya.
Popoola, Land O. Ajewolc (200 I): "Public Perceptions of Urban Forests ill lbadun,
Nigeria; Implications for Environmental Conservation". Arboricultural Journal,
Vol. 25: pp. 1-22. AS Academic Publishers
Popoola, L. and O. Ajcwole (2002):"Willingness to Pay for Rehabilitation of Ibadan Urban
Environment through Reforestation' Projects". Int. J Sustainable Dev. World
Ecology 9; 256-268.
II
UNIVERSITY OF IBADAN LIBRARY
A.1EWOLE, Opeyemi teaae/Jaur. o f Trap. For. Resources Vol. 19(2) : 2003: 1 - 12
Tobin, J. (1958): “ Lslim alion ol R elationships for Limited D ependent V ariab les’’. 
Econometrica; 26, 24-36..
UNIVERSITY OF IBADAN LIBRARY