A Preliminary Evaluation of the Panama
Women's Self-Help Construction Project
Robert Henriques Girling,
Margaret Lycette, and Nadia H. Youssef*
International Center for Research on Women
AID Contract DSAN-C-0269
The USAID Mission, Panama City
* The following ICRW staff also contributed to the design of the survey
instrument and analysis of the data used in this study: Isabel Nieves,
Mayra Buvinit, and Amy Mellencamp.
Executive Summary ...................... p. i
I. Background . . . . . p. 1
II. The Project . . . . ..... .p. 5
III. Costs and Benefits ....... ............ p. 24
IV. Replicability ........... .. .. .... .p. 35
V. Conclusions and Recommendations . . . p. 46
Appendix . . . . .... . p. 49
The Women's Self-Help Construction Project (WSHCP) of Panama was
launched in October 1981 with the goal of building 100 houses in the
municipality of San Miguelito which is adjacent to Panama City. The unique
feature of the project was that the houses were to be built entirely by
urban slum women who had no prior experience in construction.
The first phase of the project (construction of 50 houses) reached
completion in the fall of 1982. In mid-November an ICRW research team
visited Panama City to evaluate the project through a survey of project
participants and interviews with representatives of the institutions that
supported the project--Servicio Nacional de Formacion Professionel (SENAFORP),
Ministry of Housing (MIVI), and Instituto para la Formacion y el Apro-
vechamiento de Recursos Humanos (IFARHU).
Institutional support to the project consisted of training in construc-
tion skills, materials, tools, heavy equipment, and supervision. This sup-
port was forthcoming mainly due to political pressure brought to bear by
Romelia Pardo, an influential community organizer.
The actual construction phase of the project began in December 1981.
The women participants were to construct 50 houses and install all plumbing
and electrical connections, as well as work site facilities including tool
storage huts, changing rooms, showers, and a lunchroom.
Initially the women were stunned by the amount of work ahead of them.
In the end, however, they talked very positively of their experience and
the construction supervisors rated the quality of their work as equivalent
or superior to that of a professional construction crew.
ICRW survey data indicates that the women of the WSHCP were indeed
low-income women, for the most part, with low levels of education and high
unemployment rates. Forty-five percent of the women were heads of their
households with no spouse or common-law partner living with them. The
project participants, therefore, were not atypical of poor women in Panama,
yet these women managed to construct their own houses and seem to have
developed, as a result, a sense of power, greater self-reliance, and
When the costs of the project--training, materials, land, labor,
supervision, and interim financing--are compared to the benefits--housing,
effect of training on potential earnings, the positive impact of the parti-
cipatory experience, and other indirect benefits--the WSHCP certainly
appears to be a worthwhile investment. The project's benefit-cost ratio
is 1.28 with an internal rate of return of 16%. This compares to a benefit-
cost ratio of .99 for Ministry of Housing projects, with an internal rate
of return of 11.8%. For the purpose of evaluation, the effect of the train-
ing that participants received was assumed to be equal to a 10% increase in
average participant earnings. An actual study of the training effect should
be undertaken, however, upon completion of the project when the participants
again take up income-earning activities.
Perhaps one of the most important concerns of this study has been the
replicability of the WSHCP. Given the nature of the WSHCP--its development,
at least initially, through political influence--it would appear that repli-
cability will depend to some extent on continued political support for self-
help housing. In addition, continued availability of low-cost land and
materials will be important, as will be the continued viability of such pro-
jects when undertaken on a larger scale, given Panama's large housing deficit.
Finally, with limited development funds, cost recovery is a necessary,
though not sufficient, condition for WSHCP replicability.
Cost recovery cannot be achieved, however, through the use of a stan-
dard mortgage repayment scheme over 25 years at 12%. In fact, under such
a financing arrangement, only 36 % of WSHCP participants could pay for their
houses. More "creative" financing options will undoubtedly have to be em-
ployed to recover the costs of the WSHCP. Unfortunately, due to the low-
income levels of the project's participants, several of the innovative fi-
nancing options discussed in this report must involve a grant element in
order to reach the majority of WSHCP participants, thus limiting large-
Overall, the WSHCP appears to hold promise as an approach to low-
income housing solutions. Based on our preliminary findings, we suggest
that USAID continue to monitor the progress of the project; assist with the
development of community-based enterprises to allow the productive orienta-
tion developed in the project to flourish; explore mechanisms for supporting
self-help housing projects similar to the WSHCP; and consider the use of
"creative financing" options for cost recovery in such projects.
Housing is a fundamental human need. Nevertheless, throughout the
world, and particularly in the Third World, decent low-cost housing is
systematically denied to the lower income strata. In Panama the urban
poor have been relegated to marginal housing, much of it barracks which
remain from the era of canal construction at the turn of the century. A
large number of these tenement houses have been condemned. One such area
of inferior housing, much of it without electricity or water, is located
in Curundu, a virtual swampland on the banks of the Curundu River.
Here stagnant canals carry wastewater and sewage. Unemployment reaches
well above the average 19% mark cited by the World Bank for all of Panama.1
It is an area of illegal activity, i.e., prostitution, numbers games,
illegal lottery sales, theft. This is the environment and the playground
of the children of the poor. Yet Curundu is a vital community which has
been in existence for over 30 years. Its women and men have formed friend-
ships and ties over the years; they meet and talk together, buy and sell,
their children play together.
Periodically this slum area of Panama City suffers catastrophe. When
the rains come the dirt streets are awash with sewage water, houses are
flooded, and children perish in the inundated canals. Fires are frequent
in the wooden tenements, started by faulty electrical wiring or cooking
fires in uninsulated kitchen areas. One such fire occurred in September
1981 destroying the residences of more than 300 families in Curundu, and
providing a timely catalyst for the development of the Women's Self-Help
Construction Project (WSHCP).
The WSHCP was launched in October 1981 with 80 women who had parti-
cipated, during the summer of 1981, in short-term training courses for
masonry, plumbing, and carpentry organized by the Servicio Nacional de
1"Panama Special Economic Report, Metropolitan Unemployment", June 7, 1982
LAC Regional Office, World Bank.
Formacion Professionel (SENAFORP); three untrained women later joined the
project. The goal of the WSHCP was to build some 100 houses in the munici-
pality of San Miguelito, a suburb located approximately 12 kilometers from the
center of Panama City. The unique feature of the project was that the houses
were to be built entirely by urban slum women, all of whom had no prior
experience in construction and most of whom had no formal labor market ex-
perience. The women were to be provided with land, on-site supervision,
and materials for construction of their own homes. In addition, they would
receive a monthly stipend to assist with the support of their families
during the period of construction. They would be expected to pay for ma-
terials and repay the stipends after completion of the project. It was the
beginning of a bold experiment.
Construction of 50 houses, the first phase of the project, reached
completion in the Fall of 1982. At that time ICRW was asked, by USAID/
Panama, to evaluate the project under the centrally-funded "Women's Socio-
economic Participation Project". ICRW would explore questions of whether
self-help housing construction could be organized on a cost-recoverable
basis, what effects the training in construction skills might have had on
the income-generation capability of WSHCP participants, and how such effects
could be enhanced in future projects.
From ICRW's perspective, additional questions of interest related to
the socioeconomic characteristics of the project participants, how they
managed their household responsibilities during the construction period,
the quality of the housing built and, most importantly, whether the bene-
fits of the project outweighed the costs when the women's labor in the
project was appropriately valued. ICRW was particularly concerned with
the opportunity cost of the participants' labor, given past experience with
development projects that assume no cost of women's time and work.
Preparatory work for the assignment, including the development of a
questionnaire to be administered to the WSHCP participants, began in
Washington, D.C. in mid-October, 1982. Fieldwork was carried out in Panama
City during the period November 14-26, 1982.
In order to carry out the WSHCP study a two-part methodology was
employed to gain information both from the project participants and from
the institutions that provided support to the project.
Participant Information. For the purpose of interviewing the
project participants, a survey instrument was designed by ICRW, and
refined to fit the Panamanian context with the assistance of Julie Otterbein
(USAID) and SENAFORP staff. The questionnaire was to elicit information
regarding: the socioeconomic characteristics of the project participants;
any change in their income that may have occurred during and after the con-
struction project; any changes in the participants' allocation of time to-
household tasks, leisure, child care, etc.; changes in household size/struc-
ture that may have helped the participants cope with household and income-
earning responsibilities during the construction period; and the participants'
perceptions of the costs and benefits of the project, including the training
in construction skills (see Appendix for a copy of the questionnaire).
The subjects of the questionnaire consisted of 54 project participants
who attended an interview session, on the WSHCP site, in response to a
request by Sra. Romelia Pardo. Pardo is a community leader and one of the
people largely responsible for developing the WSHCP. The ICRW research
team was told that women who did not attend the session were absent because
of work responsibilities.
Three SENAFORP interviewers assisted the ICRW team in conducting the
survey of the 54 participants; 12 participants were selected at random for
more intensive interviews. Following ICRW's field work in Panama,
additional project participants were interviewed, using the questionnaire,
by SENAFORP staff; the completed questionnaires were then sent to ICRW to
be included in the project study.
Institutional Information. Pertinent staff of the institutions that
provided support to the WSHCP were interviewed regarding their institution's
role in the development of the project, the institution's specific contri-
bution to the project, the effectiveness of the training participants re-
ceived, the quality of the construction, and the construction process
overall. Information on land, labor, and material costs of housing was
obtained from the Ministry of Housing (MIVI) either through interviews or
from MIVI publications. All interviews were open-ended, but were conducted
with reference to guidelines developed for the study.
The institutions visited are: SENAFORP; MIVI; and Instituto para la
Formacion Aprovechamiento de los Recursos Humanos (IFARHU).
II. The Project
Reconstruction of the history of the WSHCP has been somewhat diffi-
cult due to the limited amount of time the ICRW research team was able to
spend with project participants and an overall vagueness among participants
regarding the timing of events. Nonetheless, through interviews with WSHCP
participants, staff of the Servicio Nacional de Formacion Professionel
(SENAFORP), and AID staff we learned that the organizational base of the
WSHCP stems from a larger community-based political organization--Las
Mujeres Torrijistas. Members of the organization provide grassroot support
to the Partido Revolucionario Democratico, the political party currently
in power. In addition, many members are "trabajadoras communitarias" who
are paid by the government to perform civic functions and work in community
development. Romelia Pardo--a strongcommunity leader in Curundu--heads the
Mujeres Torrijistas Association in the district. She has been an activist
for women for many years, and was at one point elected as Curundu's deputy
to the National Assembly. Since the late 1960s she has been campaigning
for self-help construction programs both to help solve the urban housing
problem and to improve the lives of poor urban women.
In the spring of 1981, Pardo organized a group of 105 women to form
the first Organization of Women Constructors (OWC). We could not determine
precisely how these women were chosen--whether at a community meeting or
at a meeting of Las Mujeres Torrijistas. However, they were often referred
to as "volunteers". In theory, the OWC is an autonomous group, with a
president elected from among the 105 founding members. In reality, however,
it appears that the organization is tightly bound to Romelia Pardo--though
she is not officially a member--and that Pardo may actually have appointed
the President of OWC.
Following the formation of OWC, Pardo mobilized the political sup-
port of Berta Torrijos, then director of the Instituto para la Formacion
Aprovechamiento de los Recursos Humanos (IFARHU); Pardo and Torrijos then
pressured SENAFORP to organize a special training course in basic con-
struction skills for the women of OWC. SENAFORP agreed to offer training
in masonry, plumbing, and carpentry. Of the 105 founding members of OWC,
91 enrolled in basic training. Fourteen members did not undertake the
training due to health problems and inability to forego income-earning
activities during the training period.
Training courses began in July 1981 and were offered at the SENAFORP
Center with free transportation provided. Regular teaching staff in
SENAFORP were assigned to give two-month basic courses in masonry and
plumbing. Because of an instructor's illness, however, the plumbing
course lasted for only one month. Two instructors from IFARHU taught the
Most of the women were trained in one area of specialization; only
two or three were exposed to two areas. Each participant attended a
7-hour session three times weekly; class sizes ranged from twelve to twenty
women. The training in plumbing lasted one month and the training in
construction and masonry lasted two months as opposed to SENAFORP's stan-
dard of six months of training, five days a week.
Of the 91 women who enrolled in the training courses, 86 completed
the full program; five women apparently could not afford to spend the
required time in training because of their income-earning responsibilities.
The following chart indicates the number of women trained in each
specialty and the period of time during which they received training.
Number Length of Standard
Receiving Training Training (Months) Course (Months)
Plumbing 10 4 6
Construction 51 2 6
Masonry 25 2 6
Overall, the women appear to have enjoyed the training program. They
spoke appreciatively of the patience of their instructors and felt that
participation in the program had been worthwhile. SENAFORP, on the other
hand, while supportive of the intent of the training, had strong reserva-
tions about the program for two reasons:
1) SENAFORP prides itself in promoting employment development through
its training programs; yet the goal of the short-term training of
OWC members was to enable them to build their own houses, rather
than participate in the construction labor market.
2) Staff of SENAFORP felt that because of political pressure and the
immediacy of the request to organize the training program for the
women, they fell short in the training standards provided. Pro-
gram staff are apprehensive that, if called upon in the future,
they will again be given insufficient time and resources to set
up an appropriate training program for women.
Institutional Support to the WSHCP
The completion of the SENAFORP training courses roughly coincided
with the September 1981 fire in Curundu which destroyed 300 homes. This
event provided an opportunity for Pardo, her supporters, and the women
constructors to pressure the government to grant land, construction mate-
rial and other support services to an already trained labor force to
construct their homes. In October 1981, the first Women's Self-Help
Construction Project in Panama was launched with support from three gov-
ernment institutions: the Ministry of Housing (MIVI); SENAFORP; and IFARHU.
Ministry of Housing (MIVI). MIVI is the government's public housing
construction arm. The agency provided a field engineer, construction
materials, tools and heavy equipment with the understanding that the cost
of this support would be recovered upon completion of the project.
SENAFORP. Part of the Ministry of Labor, SENAFORP is responsible
for technical and professional training in Panama and provides courses
in a variety of subjects including automotive repair, furniture building
and carpentry, masonry, sewing, and plumbing. In addition to the training
offered to the Organization of Women Constructors described in the previous
section, SENAFORP provided on-site training/supervision throughout the ten-
month construction period of the WSHCP.
IFARHU. IFARHU played an organizing and facilitating role in the
WSHCP. For seven months during the construction period the Institute pro-
vided the services of a full-time social worker, and monthly stipends of
$80 2 to each project participant. In addition, daily transportation to
and from the construction site was provided to the women for a nominal fee
throughout the entire construction period; food for lunches was provided
for the first three months of construction.
As noted in the previous section, IFARHU also contributed two instruc-
tors in carpentry during the training of the Organization of Women Con-
structors at SENAFORP.
The Construction Period
In December 1981 the WSHCP moved into the actual construction phase.
Eighty-three women, of whom 80 had received SENAFORP training, arrived on
the construction site--a hill at the northwest edge of the Torrijos-Carter
Housing Project in San Miguelito. The site had been cleared and terraced
by the Ministry of Housing (MIVI). In this first phase of the project, 50
houses were to be built on three different levels. They would consist of
23 duplexes, one triplex, and one single unit. No other site preparation
had been undertaken. The women were to construct the houses and install
all plumbing and electrical connections.
The women reported that they were stunned by the empty site, the hot
sun and the realization of the work ahead. Their SENAFORP on-site instruc-
tor was initially disappointed by the attitude of his newly-trained workers,
Note: Because the U.S. dollar and the Panamanian Balboa are on a par,
the $ symbol refers to both currencies throughout this report.
and the MIVI engineer reported that, at the beginning, he and others at the
Ministry expected the project would be "un fracaso" (a disaster).
The women's construction work involved building not only the actual
houses, but also wooden huts for storage of tools, changing rooms, showers
and the lunchroom. Initially, the women divided themselves into work groups
of eight to ten according to their specialization. As the construction
progressed, however, work could not always be done on the basis of special-
ization; for example, carpenters and plumbers often had to help in masonry.
According to one woman, "no one was allowed to stand idle," although not
every woman was required on the site at all times.
Supervision was primarily the task of the SENAFORP on-site instructor,
but he was assisted by the MIVI field engineer, a full-time site supervisor
from MIVI, and a plumbing supervisor. Other engineers paid visits on an
irregular basis to inspect plumbing and electrical work.
Construction was completed in 10 months, three months more than anti-
cipated. The MIVI field engineer expects that the fifty houses to be built
in the next phase of construction will, in fact, be completed in 7 months
because of the experience the women have gained from the first phase of
The Quality of the Construction. How does WSHCP construction compare
with the work of the MIVI-trained crews? MIVI reported that they normally
employ 80 workers to build 50 houses in about three months. The ratio of
supervisors to workers in MIVI projects is the same as in the WSHCP
(although it is important to realize that the WSHCP had the benefit of a
supervisor who was an instructor as well). The quality of WSHCP construc-
tion was equivalent or superior (the finishing was better) to other MIVI
projects, though the time spent was about three times as long as that
required by a trained construction crew. It is expected that this multiple
will fall to two during the second construction phase. The women were able
to undertake all tasks and reported no difficulty attributable to the
arduousness of the labor. For example, if a 100 lb. cement sack had to be
lifted, two women would simply work together. There were few injuries
and all were minor cuts and bruises, with the exception of a fractured
ankle which resulted from a fall from a roof.
The construction supervisors rated the women as equivalent to profes-
sional male construction workers in tardiness and absence despite the irreg-
ularity of the bus service to the construction site, child illness, and
home duties. There was mention of disciplinary problems, however, and the
sometimes difficult nature of the women's relationships with supervisors
and each other. In the end, all supervisors agreed that the project had
gone well. The field engineer reported that the work was"slow but well
done," which he considered surprising given his view that "some participants
arrived knowing absolutely nothing about the work."
Completion of the First Phase. In October 1982 the first 50 homes
were nearly completed, and the organization of women constructors met to
decide who among the group should be assigned the first 50 homes. Consensus
was reached that priority should be given to women who had worked the hardest
and who faced the greatest family difficulties. Women who are waiting for
second-phase housing showed no signs of resentment about the housing assign-
ments; two of the women who had already been assigned a house were "giving
it up" for other women in greater need.
According to the project participants and Romelia Pardo, the second
phase of the project was to begin in January 1983 and be completed by June
1983. Fifteen women would be added to the project group, and would be
trained on the job by other participants.
The Women of the WSHCP
Who were the women who participated in the WSHCP? Demographic and
socioeconomic characteristics of the surveyed participants are discussed
below. Attempts to locate women who dropped out of the project after its
initiation were unsuccessful so that characteristics of the participants
cannot be compared with those of the dropouts.
Heads of Households. As can be seen from Table 1, of the sixty parti-
cipants interviewed, 45 percent were heads of household; that is, women
who had no spouse or common-law partner residing in the household and, in
one case, a woman whose resident husband did not contribute to household
income. Half of these women had a non-resident husband or common-law
partner, however, and in some cases he contributed to household income.
Another 20 percent of participants headed their households jointly with
their spouses; both household heads contributed a similar share to house-
hold income. 3
Employment. The women of the WSHCP and members of their household
were no exception to the pattern of high unemployment that is characteristic
of low-income populations in Third World urban areas. One-third of the
women reported that they had at least one unemployed adult living in their
household. Eighteen percent reported having two or more unemployed adults
residing with them. One-third of the women also reported that they them-
selves were not working at other jobs during the WSHCP construction period.
Table 2 shows the reported work activities, during the project, of women
heads of households and of women living in male- and jointly-headed house-
holds. Forty percent of women heads of households were not working during
the project, but a full 47 percent were working for pay. Thirty percent of
the other women were not working and only 36 percent were working for pay.
Fewer women who were heads of households reported being housewives than did
other women and the difference approached statistical significance.
(X2 = 12.17 df = 6; .10p-1.05).
Income. Table 3 shows the distribution of household income during
WSHCP construction for women heads and for other women in male and jointly-
headed households.4 The majority of the participants had incomes of less
than $255 monthly and thus fall into the two lowest deciles of the Panama
income distribution. Women heads were somewhat poorer than other women:
Although yet undocumented, joint headship is probably quite prevalent
in low-income urban areas of Latin America.
4This figure includes the $80 monthly stipend.
This figure includes the $80 monthly stipend.
PARTICIPANTS' HOUSEHOLDS BY TYPE
TYPE / No. %
Woman-Headed 27 45
Male-Headed 21 35
Jointly-Headed 12 20
TOTAL 60 100
1/ Head(s) of Household is (are) the residents) reported
as contributing the largest share to household income.
PARTICIPANTS' REPORTED WORK ACTIVITIES
BY TYPE OF HOUSEHOLD
TYPE OF HOUSEHOLD
WORK ACTIVITIES JOINT HEADED
No. % No. %
Not working 10 30 11 41
Community Workers 9 27 9 33
Housewives 7 21 3 12
Domestics 2 7
Laundresses 2 7
Construction workers 3 9
No response 4 12
TOTAL 33 100 27 100
the majority of women heads (63 percent)had incomes below $134 per month
while only 15 percent of other women reported such incomes (the difference
is marginally significant; X2 = 6.2, df =4; .10JpJ.05; one tailed test).
Education. All but three women interviewed reported having some
primary schooling. In fact, 27 percent of the women had completed some
primary education; 33 percent had completed primary school; and another
33 percent had some secondary schooling. (See Table 4 ). Only seven women
heads of households had some secondary schooling and, overall, women heads
had slightly lower levels of educational attainment than women in male-
headed households or in jointly-headed households; this difference was not,
however, significant. (X2 = .78 df = 1).
Age. The ages of the women participants ranged from 19 to 62 years
old. Over 65 percent of the women, however, were less than 40 years old
and over 88 percent were less than 50 years old. (See Table 5 ). The
median age for the participants was 34 years. There was no significant
difference in the age distribution of participants in jointly- or male-
headed households, versus those in woman-headed households. (X2 = 1.45 df=3).
Most women in the project, therefore, were in their prime childbearing years.
Initially we had hoped to explore whether women with young children had found
it difficult to participate in the project due to child care responsibilities.
While it was not possible to interview those women who dropped out of the
project, in order to see if child care problems were a major factor in their
attrition, it was possible to check on the number of women with young chil-
dren who remained in the project.
Children. As Table 6 indicates, slightly more than half of the women
interviewed indicated that they had children under six years of age living with
them. There were no significant differences in the numbers of children
under six years old in woman-headed households versus male- or jointly-
headed households (X2 = .50 df = 1).
Coping with Household Responsibilities. How did the WSHCP women cope
with eight or so hours of construction daily, other income-earning work,
REPORTED HOUSEHOLD INCOME BY TYPE OF HOUSEHOLD
INCOME ALL HOUSEHOLDS MALE-AND JOINT-HEADED WOMAN-HEADED
($) No. % No. % No. %
0 134 22 31 5 15 17 63
135 255 28 42 19 57 9 33
256 365 7 14 6 19 1 4
366 475 2 5 2 6 0 0
476 600 1 4 1 3 0 0
601 + 1 4 1 4 0 0
TOTAL 60 100 33 100 27 100
EDUCATION OF PARTICIPANTS
EDUCATION ALL PARTICIPANTS WOMEN-HEADS OF HOUSEHOLDS(WHH)
ATTAINED No. % No. %
None 3 5 1 4
Some Primary 16 27 9 33
Completed Primary 20 33 9 33
Some Secondary 20 33 7 26
Completed Secondary 1 2 1 4
TOTAL 60 100 27 100
AGE DISTRIBUTION OF PARTICIPANTS
AGE GROUP ALL PARTICIPANTS WOMEN-HEADS OF HOUSEHOLDS(WHH)
(YEARS) No. Cum.% No. Cum. %
15-19 3 5.0 1 3.7
20-24 8 18.3 4 18.5
25-29 11 36.7 4 33.3
30-34 10 53.3 3 44.4
35-39 7 65.0 4 59.2
40-44 6 75.0 3 70.3
45-49 8 88.3 4 85.1
50-54 2 91.7 1 88.8
55-59 3 96.7 2 96.2
60-64 2 100.0 1 100.0
TOTAL 60 27
PRESENCE OF CHILDREN UNDER SIX YEARS OLD
IN ALL HOUSEHOLDS AND THOSE HEADED BY WOMEN
CHILDREN--5 YEARS ALL PARTICIPANTS WOMEN-HEADS OF HOUSEHOLDS(WHH)
No. % No. %
No Children 28 47 14 52
Children 32 53 13 48
60 100 27 100
household chores, and child care responsibilities? Participants seem to
have relied somewhat on other people to help them with household tasks
during construction. Two-thirds of the participants reported having the
help of another household member (see Table 7 ), while few participants
reported having the help of non-resident relatives, friends, or other
For most women, however, help was not very frequent, and seems to
have been specific to certain household tasks. High proportions of the
women never received help, from either children or spouse, with some
household tasks. For example, no women reported having help with shopping.
Only in child care, carrying water, and washing clothes did a substantial
proportion of the women--18 to 33 percent--receive daily help, and this
help was from children rather than spouses. (See Table 8).
If these responses are valid, most women must have had to sacrifice
leisure time in order to work on the housing project. This may have been
the case even for those who had their children's daily help if such help
is usual and chores did not increase substantially during construction.
Unfortunately, our survey yielded no information with which to test this
The Women's Experience. The women talked very positively of the
training and the construction phase, and did not indicate any problems in
working together. They felt that the training sessions and construction
work fit into the normal routine of their day. They continued to rise at
4:30 a.m. to prepare food for the day; they left home at 6:00 a.m. and
returned after 3:30 p.m., some to take up a job for a few hours, others
to engage in informal trade. Saturday were reserved for washing clothes,
cleaning house and other household tasks.
The women claimed they lost no hours from entertainment because they
had seldom experienced such activities. In fact, some said that the first
time they enjoyed "entertainment" was during the construction job.
Child care was not considered problematic. It is,apparently, common
practice for them to leave children over age 5 at home alone. The women
HELP DURING CONSTRUCTION AS REPORTED BY PARTICIPANTS
PERSONS WHO HELPED %
Household Members 67
assured us that neighbors "look in" on their children at different times
during the day. Girls are taught to cook at age 6 and heat prepared food
for themselves and younger siblings in their mother's absence. In three
cases a relative came to stay with the children during the construction
period and one woman hired child care help at a cost of $30 per month. A
community child care center was available but not used.
A relatively high number of the women interviewed were "trabajadoras
comunitarias" who continued to receive their regular salary ($125) through-
out the construction period since government employees can be reassigned
to other projects, continue to receive their salaries, and preserve their
positions. Through Romelia Pardo and Berta Torrijos, almost all the traba-
jadoras comunitarias in the project were able to secure this reassignment.
The other women derived income from informal sources and it is inconceivable
that their activities were not affected by the time spent on the project.
However, they made no reference to any economic losses incurred. Some
women, including trabajadoras comunitarias, engaged in "trade" when at
home--an activity which apparently was not affected by the work schedule
of the construction project. Two women continued to work full-time jobs
in the evenings after returning from the construction site.
Of the women who began work on construction, only three dropped out
during the course of the ten-month period--reportedly due to medical and
financial problems, and, in one case, death of the participant.
Two women from outside Curundu were brought in as replacements. Both
women were voted in by the group because of their participation in relief
work in Curundu during a flood. The project participants did not appear
to mind the inclusion of "outsiders."
At this point it may be worthwhile to specify some of the issues that
have been raised by the WSHCP. First is the issue of whether or not the
project has been worthwhile and, therefore, should be replicated. Section
III of this report addresses this issue, discussing the costs and benefits
of the project. The effectiveness of the training provided through the
project enters the benefit calculation in terms of its contribution to
improving the income-generation potential of the participants.
Next, we must ask whether the project can be replicated. Section IV
discusses this issue in terms of whether the leadership and sense of com-
mitment among project participants exhibited in the project can be dupli-
cated; whether there are viable cost recovery options for projects of this
sort; and whether such projects can be replicated on a larger scale.
It is not, however, within the scope of this report to assess the like-
lihood of continued support for self-help housing projects in Panama.
Rather USAID will have to further explore this question when,and if, support
for self-help housing is considered.
III. Costs and Benefits of the Project
Has it made sense for the women of the WSHCP to construct their own
houses, or would they have been better off purchasing already built low-
income housing? This section of the report examines the costs and the
benefits of the WSHCP both in absolute terms and relative to the low-income
projects undertaken by the Ministry of Housing.
The total costs of the Women's Self-Help Construction Project consist
of indirect costs--training--and direct costs--materials, land and infra-
structure, labor and supervision, and interim financing. All costs have
been estimated using, whenever possible, several sources or methods of
estimation in order to improve reliability. In Table 9, three estimates
are presented for each cost category--a high estimate, a low estimate, and
a "most likely" estimate. The "most likely" estimates are used in calcula-
tions of the project's benefit-cost ratio. However, since the amount of
stipends paid to the WSHCP participants is below the most likely opportunity
cost of their labor, and the participants cannot be expected to repay more
than they actually received for their labor, the recoverable cost of the
WSHCP is assumed to be equal to the "most likely" estimate of total cost
per unit,less supervision and the difference between actual stipends paid
and the most likely imputed value of labor on the project. Supervision
costs are not considered recoverable since they are borne through the taxes
used to pay SENAFORP and MIVI staff salaries.
Costs of the WSHCP units may be compared to the actual costs of
MIVI core unit projects and to the estimated costs of MIVI projects
if built on the type of land and lot sizes.similar to those used in
Training Costs. Training costs were estimated from SENAFORP data on
materials cost per participant and instructor salary information. The cost
estimates range from $100 to $160 per unit. It is important to note,
however, that these costs are borne to some extent by the salaried partici-
COSTS PER UNIT ($)
(ESTIMATE, REDUCED LOTS)
Surveying and Foundation
Plumbing & Fixtures
Land and Infrastructure
Memorandum Item: Training
e MIVI reports no supervision costs. This figure
reflects profits per unit.
N.A. Not applicable
pants. Training costs in Panama are financed from a tax on salaries and
profits of 1.5% and training is available as a government service. The
training cost estimates, therefore, are noted as a memorandum item and are
not included in the calculation of the total cost per unit.
Material Costs. The estimated costs of material, including founda-
tion, ranges from $1,952 to $2,271 with a most likely estimate of $2,196.
These figures are 8% higher than the costs of equivalent MIVI units, even
though the materials for the WSHCP were supplied by MIVI at cost. The
higher cost of WSHCP materials seems to be attributable to a higher cost
of foundation work in the project. Foundation work was the first construc-
tion task undertaken in the project and because the participants were
inexperienced they took longer to complete the task than MIVI crews require.
Thus, the heavy equipment used in foundation work was on the site, i.e.,
being rented, for longer than is usual in MIVI projects, boosting the
materials cost to the WSHCP.
Land and Infrastructure Costs. Land and infrastructure costs have
been estimated based on MIVI costs and adjusted for the smaller lots and
quality of land used in the WSHCP.
Labor and Supervision Costs. Perhaps the estimate most difficult to
make, from a methodological point of view, is that of labor costs. It was
initially intended to value the labor of the participants in terms of the
opportunity cost of their time, i.e., the amount of money that the women
would have earned had they been engaged in their normal income-earning
activities. This amount was to be measured by estimating the actual devia-
tion in family income during construction from normal family income. Ad-
ditionally, costs incurred as a result of a woman's work on the project
rather than in household tasks--such as increased food costs because the
household had to buy more already prepared foods--were to be factored into
As it turns-out, family incomes actually increased overall during the
construction period because, as mentioned previously, a high proportion of
project participants were "trabajadoras comunitarias" who retained their
$125 monthly salaries throughout the project period. In addition, project
participants received an $80 monthly stipend from IFARHU. Finally, many
women continued to carry out their normal work after the construction day,
on weekends, or occasionally by doing their normal work in lieu of going
to the construction site. Obviously, these women must have reduced the
time they devoted to household tasks or leisure and the costs of this fore-
gone time should be considered. Unfortunately, the responses to the survey
questions dealing with such costs are few and appear to be somewhat unreliable.
Labor costs, therefore, have been estimated as follows:
low estimate: total amount of stipends that will have been paid
by IFARHU upon completion of the entire 100-unit project, $91,400
for a per unit cost of $914.
high estimate: the total amount of IFARHU stipends plus 75% of
the total salaries paid to the "trabajadoras comunitarias" which
we regard as the upper limit of the value of their time allotted
to construction work. This results in an average unit cost of
$1197 for both phases of the project.
most likely: an estimate assuming a $3.50 daily shadow price of
unskilled labor calculated by MIVI,6 resulting in a unit cost of
$1137.50 (average for both phases of the project).
The best estimate of supervision costs, based upon the salaries of the
supervision personnel provided by SENAFORP and MIVI is $140 per unit. It is
important to note that part of the supervisory task was training. In view
of the abbreviated training given to the participants a portion of this
could equally well have been included in training costs. In any event our
Since 80 of 83 participants accepted stipends to build only 50 units, and
in the next phase of the project 17 new women will join and receive sti-
pends, accuracy required that an average per unit cost of labor be calcu-
lated as follows: total stipends paid by IFARHU in the first phase of the
project, plus stipends to be paid in the second phase (assuming only 6
months of support to the project since construction time is expected to
be reduced in the second phase), divided by 100 units.
MIVI, "Seguimiento a los Proyectos Roberto Duran y Torrijos Carter".
view is that this cost should not be regarded as recoverable because, as
with training costs, SENAFORP and MIVI salaries are financed through taxes
which are already borne by the salaried project participants.
Interim Financing Costs. Interim financing is the cost of the use of
capital during the construction period. Our estimates of this cost are
calculated assuming a 12% interest rate for a full year on the total value
of project land and infrastructure and on the average monthly amount of
Total Costs. The range of total costs of the WSHCP is from $4856 to
$6603. The most likely estimate is $5844, $400 below the MIVI cost of
$6244 for its typical unit. Despite the fact that WSHCP material costs,
labor, and interim financing expenses were higher than MIVI's, the total
cost is lower, since the costs of land and infrastructure at the WSHCP site
were substantially less due to the small area of the site and marginal
quality of the land. The most reliable estimate of recoverable cost is
$5480. This is derived by reducing the total cost (which does not include
training) by supervision costs as well as the difference between the imputed
labor costs and the amount loaned to the participants by SENAFORP. (Note:
It is expected that in phase two of the project costs will decline by about
$300 per unit due to a shortened construction period.)
The benefits of the Women's Self-Help Construction Project may be seen
from a variety of perspectives and points of view. Some are clearly eco-
nomic, while others are more broadly socioeconomic. In this section the
main benefits of the WSHCP are described and, whenever possible, estimates
made of their economic value. We have analyzed the benefits both on the
basis of what the participants told us were the main benefits as well as
what micro- and macro-economic theory would suggest.
Direct benefits of the project include the housing built in the pro-
ject; the value of the improved environment and services such as water,
drainage, electricity; the training that project participants received; and
the benefits of the participatory experience itself, such as improved self-
esteem, greater sense of community, etc.
Indirect benefits of the project include the potential contribution
of the participants' construction skills to the repair and expansion of the
houses of friends and to community building projects; transmission of skills
to future project participants--the incoming group of 17 new participants
for the second phase of the WSHCP will be trained by the first phase parti-
cipants; and intergenerational effects--the skills transmission and environ-
mental improvement will have a positive effect on succeeding generations.
Housing and Environment. The annual benefit (E) of WSHCP housing,
along with improved environment and water, electricity, and other services
is estimated as the annual payment required to amortize the estimated cost
of 50 comparable MIVI units with reduced lot size ($5154/unit) at a 12% rate
of interest over 25 years--$32,470.
Almost all project participants felt that housing and improved environ-
ment were the main benefits of the project. The WSHCP site, bordering the
Torrijos-Carter project area, was thought to provide a healthier, safer and,
in general, far better environment for children than Curundu where children
would be exposed to criminal activities and would be encouraged or pressured
to participate in such activities.
Training and Potential Effects on Income Generation. The SENAFORP
training that the WSHCP participants received was not intended to result
in construction sector employment and did not prepare women to be con-
struction workers. It did, however, prepare the women to expand houses
and core units; organize a construction cooperative; and/or work as
independent contractors on small construction jobs. 7
7 The alleged agreement between construction unions and the government of
Panama to exclude women from the construction industry does not, apparently,
apply to small contractors or women in cooperatives.
The women of the WSHCP perceived the benefit of training principally
in terms of the personal value of learning a skill. Because only the first
phase of the project has been completed, the participants do not yet seem
psychologically geared toward using their skills for employment other than
to finish the second phase of the WSHCP.
When the women did think about income-generation activities and work op-
portunities many were interested in community-based enterprises and
cooperative forms of production managed by and for the community. Those
who specialized in plumbing and in carpentry showed the most interest in
applying their skills to generate income in cooperatives or as independent
contractors; women trained in masonry were less interested and found con-
struction work tiring.
At the time of our interviews, none of the women had attempted to earn
income with their construction skills. It is impossible, therefore, to
value the training benefit with a high degree of reliability; ICRW hopes
to return to Panama following completion of the project's second
phase in order to better assess the impact of the construction training.
In the meantime, given both the training of the participants and their
current and expected on-site experience it seems reasonable, if not con-
servative, to assume a 10% increase in participants' average earnings as
a training effect (T) equal to $9000 per annum.
The Participatory Experience. A review of participant and institu-
tional interviews indicates that significant processes that are unquan-
tifiable have taken place amongst the participants in the WSHCP. We believe
that these are "benefits" that have been generated and strengthened in
large part by the participatory experience itself and which are of signi-
ficance beyond the immediate construction effort.
Improved Self-Perception: The experience in mobilization and group
action in demanding and obtaining the right and capability to actually
build their own homes, has developed in the participants a sense of power,
greater self-reliance, pride in themselves, and dignity. Once people
acquire these qualities they can more readily seek out opportunities for
improved personal and economic welfare. The WSHCP experience may have
generated a sense of control that will unleash the participants' ability
and desire to direct their lives in new and meaningful directions. Cer-
tainly some evidence of this was given during in-depth interviews. Women
repeatedly said that 'nobody believed we could build our homes and we
showed them that we did...'; 'now that we have built our own house, there
is so much else that we are going to be able to do...'. Not one woman
complained of the time spent and the energy expended in either the training
or construction stage of the project; and several women reported a per-
ceived improvement in health as a result of the project.
The IFARHU social worker shared these perceptions; in her words the
project experience had "transformed" the women and succeeded in bringing
out in them a sense of consciousness, social responsibility, self-reliance,
and pride. Her characterization of the women's experience was shared by
the SENAFORP on-site instructor who was most closely involved with the
women during the construction phase.
Sense of Community/Solidarity: Additionally, the WSHCP may have
fostered a sense of group solidarity. There are indications that the
project participants have developed a consciousness of a larger whole
whose welfare is every individual's concern. The women faced common
problems as a group and found solutions collectively, leading to greater
self-assurance and pride in the group. The strength of group feeling has
been most tangibly demonstrated through:
the participation of all the women in the building effort for the
community, without knowing to whom the first houses would be assigned;
the collective agreement, at the end of the first phase of the con-
struction project, to assign the first 50 houses to those women who
had worked the hardest and had serious family problems:
the strong commitment expressed by some women, including those who
had already been assigned housing, to complete the second phase of
the project; and
the expressed interest of the WSHCP participants in cooperatives and
How much of this consciousness can be attributed to the participatory
aspect of the project and how much to the experience of the participants
prior to the construction project is unknown. Most participants knew at
least one other participant before the project, although only a few women
identified themselves as having been members of a social/community organi-
zation before participating in the housing project. It may be that a
"group feeling" was actually fostered in the Curundu district, prior to the
project. In any event, the effects of the participatory experience cannot
be quantified and should merely be kept in mind as a positive aspect of the
Indirect Benefits. The indirect benefits of the project have been
estimated as follows:
The potential contribution of construction skills in the community (C)
is equal to $6000 per year, based upon estimates developed in two 1981
studies of self-help housing.8 This benefit refers to the additional
housing value that would result from future building and expansion
of the housing units following the completion of the project and using
the skills acquired during the project.
The transmission of skills to future project participants (S) is equal
to $3000, the estimated cost saving due to the training of the new
participants in the second phase of the project by the first phase
participants. This is a first year benefit only.
Intergenerational effects could not be quantified.
8 MIVI "Seguimiento a los Proyectos Roberto Duran y Torrijos-Carter", March
1981, pp. 47-49; "A Study of the Progressive Development of Three Low-
Cost Housing Projects in Panama," AID, Office of Housing, Occasional Paper
Series, Spring 1981, pp. 50-51.
WSHCP MIVI lot size
(50 Units) (50 Units) (50 Units)
Total Project Costs ($)
Materials 109,800 102,000 102,000
Land & Infrastructure 97,500 152,000 97,500
Labor 56,875 36,600 36,600
Supervision 7,000 7,100 7,100
Financing 21,000 14,500 14,500
Total 292,175 312,200 257,700
Total Annual Benefits ($)
(E) Housing & Environment 32,470 39,243 32,470
(T) Training a/ 9,000 -
Participatory Experience b/ + -
(C) Contribution to Community Construction-/ 6,000 -
(S) Transmission of Skills d/ 3,000 -
Intergenerational Effects +
(PV) Present Discounted
Value of Benefits d/ 374,990 307,787 254,566
Benefit/Cost Ratio/ 1.28 .99 .99
Internal Rate of Return 16% 11.75% 11.82%
a/ Based upon 10% increase in average participant annual earning.
b/ While it is likely that the value is positive, there is no basis for
making quantitative evaluation of these items.
c/ Based upon survey of self-help activities in Torrijos-Carter.
d/ First year benefit only. 25
25 B= Benefits = Et+ Tt +
e/ Bt t=1
S V= (l+i)t i =12% per annum
t = period in which benefit i
Ct + St
f/ Benefit/Cost Ratio = PV-- Total Project Costs.
Total Benefits. We calculate the present value of the WSHCP's benefit
stream, PV, as follows:
PV= E t
where Bt= E Et + Tt + Ct + St
i = 12%
E = annual benefit of housing and environment
T = training effect
C = contribution of construction skills in the community
S = value of skills transmission to new participants
Benefit-Cost Ratio. According to our analysis, the total present
value of benefits is $374,990. This compares with project costs of $292,175
yielding a ratio of benefits to costs of 1.28. (See Table 10). The internal
rate of return which equalizes the cost and benefit streams is 16%.
How do these figures compare to the alternative of construction by MIVI
with the women simply purchasing their houses? Total project costs would
be $312,200; annual benefits would be reduced to $39,243, the value of the
units amortized at 12% over 25 years, for a present value of $307,787. The
benefit-cost ratio would be .99 with an internal rate of return of 11.75%.
Of course, MIVI units are more costly than WSHCP units because they are
built on larger lots and higher quality of land. If we assume that MIVI
would be willing to reduce lot size and quality, the costs of MIVI units
could be reduced to $5154 per unit.9 Total MIVI project costs would be
$257,700; total present value of benefits would be $254,566. In this case
the benefit-cost ratio is again .99 due to the reduced value of land and
lot size. The internal rate of return is 11.82%, and WSHCP units again
9Some recently built MIVI units have reduced lot sizes.
Some recently built MIVI units have reduced lot sizes.
We come now to perhaps the most important concern of our study:
the replicability of the WSHCP. It has been shown that the benefits of
the first phase of the project outweigh the costs and that the internal
rate of return for the WSHCP is higher than that of comparable MIVI housing
developments. Clearly, then, the WSHCP is a "successful" project, not
only from the point of view of its participants but also in terms of the
economies of the project and, therefore, should be replicated. Given that
the project should be replicated, we must ask if the project can be repli-
cated. There would seem to be at least four requirements for replicability:
political support and leadership; availability of land and resources such
as construction materials, heavy equipment, credit, etc.; large-scale pro-
duction while maintaining (or improving) the internal rate of return; and
Political Support and Leadership
Given the nature of the WSHCP, its development--at least initially--
through the use of political influence, its roots in the "Mujeres Torrijistas",
and the strong leadership of Romelia Pardo, we must conclude that the project
was to a major degree motivated by the fortuitous intersection of a unique
set of social conditions with a political movement. Replication of the pro-
ject, then, may require duplication of the leadership and political/public
sponsorship given the WSHCP. As mentioned earlier, there are indications
that political support for the WSHCP may be on the wane. It remains to be
seen whether this diminished sponsorship will apply to the WSHCP only, be-
cause of its association with Berta Torrijos, or to self-help housing projects
in general. In the latter instance, replication of the WSHCP will be dif-
ficult if not impossible.
Availability of Land and other Resources
The availability of low-cost land, financial support in the form of
stipends, transportation services,and materials supplied on credit and at
cost to the WSHCP certainly was crucial to the success of the project. With
the rising costs not only of urban land, but also of suburban land, and
competing, urgent demands on government and financial resources in Panama
and in most Third World countries, it will be increasingly difficult to
maintain the availability of these critical resources in order to replicate
the WSHCP. This constraint to replicability is not unique, of course, to
the WSHCP or indeed to any one project. Allocation of resources to any
project or sector implies a denial of resources elsewhere. Our study indi-
cates that self-help housing in Panama compares favorably to government low-
cost housing schemes, with a given amount of resources devoted to the
housing sector. Whether self-help housing projects, however, should take
precedence, in the competition for resources, over industrial development
projects, education projects, nutrition programs, etc., will depend on the
importance attached by society to the various sectors. In the final ana-
lysis, therefore, replicability of self-help projects will require a good
deal of political will.
Panama's housing deficit is large; recent data prepared by the Ministry
of Housing shows a deficit of 185,000 units for the entire country including
80,000 in the urban areas. To what extent could self-help housing projects
such as the WSHCP be replicated on the large scale required to reduce, even
partially, that deficit? Unfortunately we cannot answer this question based
on the study of one not-yet-complete project. On the other hand, we can
suggest that one major requirement for replicating the WSHCP on a larger
scale, while maintaining a benefit/cost ratio and internal rate of return
higher than those of MIVI, would be the duplication of the group feeling
and community bonds evident during WSHCP participants. It is our impression
that these feelings of solidarity and commitment were beneficial to the
project in that they inspired the participants to work hard and consistently
on the construction.
It appears that the participants' group feeling began to develop at
the time that they attended the SENAFORP training program and was solidified
during the ten-month housing construction period. The participants' feeling
that they were building houses for their community and not just for them-
selves seems to have been a significant factor in the strengthening of bonds
within the group. In-depth interviews revealed that most participants had
known another member of the group before the initiation of the project.
However, very few identified themselves as belonging to a social or community
organization prior to the housing project.
Interestingly, group activities seem to have been restricted to the
training and construction sites. Women shared transportation to the con-
struction site, and engaged in communal cooking there. However, in none
of our talks with the women were there any indications that there existed
arrangements for mutual help in other spheres of the women's lives, such
as child care or income-earning activities. This may bode well for repli-
cability insofar as it indicates group bonds may be developed fairly easily
on a project site; project developers need not worry that non-project
group activities must take place in order to ensure a sense of group re-
sponsibility for project activities.
The concept of cost recovery is based on the principle that funds
expended by agencies in undertaking a project should be repaid or recovered
so that these funds can then be recycled to other, similar projects or used
to expand an entire development program. That is, given limited development
funds, cost recovery is a necessary, though not sufficient, condition for
project replicability on a meaningful scale.
In typical housing projects, houses are built and then sold to those
who can afford them. Affordability is sometimes enhanced through the use
of "cross-subsidy" schemes whereby 'profits' derived from the sale of higher-
cost housing are used to reduce the price of the basic housing units in-
tended for low-income families. Mortgages are used to recover direct
project costs; taxes recover many indirect project costs.
In the case of the WSHCP, the circumstances are atypical. No consid-
eration, prior to the project, was given to the income levels of the women
who participated in building their houses. Now we must ask whether these
women can afford to pay for their houses: can cost recovery be accomplished
in the WSHCP?
The indirect costs of the WSHCP--training and supervision by SENAFORP--
will be recovered through the Panamanian tax on profits and earnings. While
there are several different methods of recovering direct costs of the pro-
ject, all are variants of the mortgage mechanism. The ideal method will be
one that does not impose high administrative burdens or risks while maxi-
mizing the number of project participants who can afford to buy their houses.
Maximum affordability is, of course, key to the success of cost recovery.
Table 11 shows the monthly income range and the range of affordable
monthly housing payments for the first six deciles of the income distri-
bution in Panama. We calculate affordability on the assumption that
families can devote 25% of income to housing, except for the poorest decile
who can afford no more than 10% of their income. The figures in Table 11
reflect that assumption.
Table 12 shows the percentage of WSHCP participants in each income
decile along with, again, the average affordable monthly housing payment
for the decile. It is immediately apparent from the table that nearly
one-third of WSHCP participants are quite deprived, falling into the lowest
income decile; another 42% of participants are in the second income decile;
and a full 96% of project participants have incomes lower than those of
50% of the population of Panama. If the WSHCP participants are to afford
their housing, some of the more "creative" cost recovery options discussed
below will have to be employed.10
Standard Payment Option. The standard mortgage currently used by
MIVI involves a 12% interest rate and 25-year term mortgage. Given these
terms, the required monthly payment for amortizing a loan of $5480, the
"most likely" recoverable cost of WSHCP units, is $57.54 (excluding insur-
ance.) (See Table 13.) Under the traditional payment structure, therefore,
only about 36% of the WSHCP participants would have the minimum monthly
income required--$230--to meet the monthly payment; only women from house-
holds with incomes in approximately the 19th percentile and above could
1 A study by MIVI in 1981 calculated that a payment of $50 a month or
more would be beyond the capacity of 70% of Panama's population (Nuevo
Tivoli, "Analisis Socio-Economico Proyectado al Programa de Renovacion
Urbana." MIVI. October 1981, p. 33.)
PANAMA INCOME DISTRIBUTION AND
AFFORDABLE HOUSING PAYMENTS
% OF POPULATION INCOME RANGE ($) AFFORDABLE MONTHLY HOUSING PAYMENTS($)
WSHCP PARTICIPANTS' INCOME DISTRIBUTION AND AFFORDABLE HOUSING PAYMENTS
MONTHLY INCOME % OF WSHCP PERCENTILE OF PANAMA MINIMUM AFFORDABLE
($) PARTICIPANTS INCOME DISTRIBUTION HOUSING PAYMENT ($)
= $ 72,
STRATEGIES FOR COST RECOVERY IN THE WSHCP
COST RECOVERY STANDARD REDUCED REVOLVING FUND/ SERVICED COMMUNITY NEGATIVE
OPTIONS: REPAYMENT INTEREST RATE DEFERRED REPAYMENT LOT PAYMENT AMORTIZATION
Amount to be
Amortized ($) 5480/unit 5480/unit (a) 914/unit
(b) 4566/unit 2050/unit 548,000/100 units 5480/unit
Interest Rate 12% 9% 12% 12% 12% (a) actual: 12%
(b) base of payments:
variable, beginsat 5%
Term (Years) 25 25 (a) years 1-3 25 25 Variable
(b) years 4-22
Payment $57.54 $45.48 (a) $30.34 $ 21.52 (a)$4361 individual $ 31.78 initially
(b) $48.85 (b)$1393 community
included/ 36 52 72 79 100 71
Grant element None $1145/unit (a) support of revolving None technical assistance None
fund for 3 years/ with community
(b) grace period enterprises
1/ Present value of grant = $ 186,092. Note
indefinitely. (See Table 14).
however that 100 units per year may be financed by the fund
2/ Normal incomes of participants, i.e., excluding $80 monthly IFARHU subsidy.
Reduced Interest Rate Option. If the interest rate charges were to
be reduced to 9% the situation would improve appreciably. In this case the
monthly payment required would fall to $45.48, allowing 52% of the parti-
cipants (those in the 15th percentile and above) to own their homes. Never-
theless, 48% of the sample of project participants who need and want hous-
ing, and were willing to work to obtain it, would still be excluded.
Of course, a 9% interest rate would be well below the market interest
rate; the present value of the subsidy required to reduce the rate to such
a low level would be $1145 per unit and thus would impair the wide-scale
replicability of the project.
Revolving Loan Fund/Deferred Repayment Option. Another option for
cost recovery would involve dividing the unit cost of WSHCP housing into
two portions: $914 in direct labor costs (stipends) and $4566 in materials,
land and capital costs. The $914 would be handled as a revolving loan to
be repaid at 12% over the first three years of a 25 year term; a three-year
grace period would apply to the remainder of the total principal which
would be repaid at 12% in years 4 through 25. This means that $914 must be
amortized over 3 years and $4566 must be amortized over 22 years. The re-
sulting required monthly payments at a 12% rate of interest are $30.34 per
month during years 1 to 3, rising to $48.85 per month during years 4 to 25.
What are the advantages of this approach? First, the revolving loan
concept may help provide continuity in the self-help construction process.
As one group of participants complete their homes their repayments will
begin to be available to fund a subsequent group of constructors. The
details of the capitalization required for such a revolving loan fund,
assuming fixed costs of bad debt and operation, are noted in Table 14.
The fund would be self-supporting after three years.
Second, the deferred repayment aspect permits a wider range of parti-
cipants to pay their housing costs during the first three years of the
mortgage term. During this period, methods of raising household income
REVOLVING LOAN FUND FOR LABOR (STIPEND) COSTS
OF 100 HOUSING UNITS PER YEAR (IN 1982 DOLLARS)
YEAR 1 YEAR 2 YEAR 3 YEAR 4 YEAR 5
for stipends $ 91,400 $ 91,400 $ 91,400 $ 91,400 $ 91,400
Bad debt and
Operating Costs a/ $ 15,000 $ 15,000 $ 15,000 $ 15,000 $ 15,000
this period $ 36,408 $ 72,816 $109,224 $109,224
Net of funds loaned,
bad debt, and
operating costs $106,400 $ 69,992 $ 33,584 + $2824
this period b/ $106,400 $ 69,992 $ 33,584 0 0
Fund Balance 0 0 0 $2824 $5648 C
a/ These costs are here assumed to be fixed. The funds' cash flow will of
of course change if that assumption does not hold.
b/ Present value of capital required for the fund in years 1 through 3 =
$ 186,092. (Discount rate= 12% p.a.)
c/ Fund balance continues to increase by $ 2834 per annum.
can be developed; for example, a number of the WSHCP participants mentioned
that they would like to form construction and repair cooperatives while
a few hoped to open bodegass" and bakeries to earn income. A three year
period of low monthly payments would provide a watershed period during
which the participants might build their incomes to the point where they
could afford the higher payments required in later years of the repayment
schedule. Under this method of cost recovery 72% of our WSHCP sample could
afford housing, compared to 36% under the standard repayment option.
Nevertheless, there is a risk associated with this approach: the in-
comes of some participants may not grow sufficiently to enable them to
afford the 60% increase in the required monthly payments after the third
year of the repayment term. These participants may then have to default
on their mortgages.
Serviced Lot Option. Permitting low-income participants to build not
a house but a serviced lot could reduce the cost of housing from $5480 to
$2050. Low-income participants, however, would be given possession of a
serviced foundation in lieu of a 25.2 m2 core unit. Under this option
approximately 79% of the WSHCP participants could afford "housing" at a
minimum monthly cost of $21.52. This option would reach those in the 8th
percentile of the income distribution and above. However it is not clear
whether low-income women would readily participate in building the higher
cost units for others when they themselves will receive only serviced
lots. That is an empirical question requiring further exploration.
Community Payment Option. A cost recovery option that would include
all project participants, inspired in part by our observations of the spirit
of sharing and communal responsibility shown by the participants of the
WSHCP, involves an income-pooling arrangement. If we aggregate the cost
of the project units--100 units upon completion--the monthly payment for
the entire project, amortized over 25 years at a 12% interest rate, is
$5754. This amount could be paid through both individual payments and a
group payment. Individual payments would be made out of each participant's
household income according to an affordability criterion, e.g., 25% of
household income up to a $65 maximum. Individual affordable payments would
generate approximately $4361 leaving $1393 to be earned each month by com-
munity-owned enterprises, or $16,716 annually.
The drawback to this option is the large amount of money that would
have to be generated each year through community enterprises. This raises
an interesting possibility for involving AID's IIPUP Program to assist WSHCP
participants in developing community enterprises that could generate the
annual payment required.
Negative Amortization Option. Under the negative amortization option
a 12% market rate of interest would be applied to loan balances. Payments,
however, would initially be calculated on the basis of 5% interest rate,
allowing the inclusion in the project of low-income households. Monthly
payments would initially be $31.78. The rate of interest used to calculate
payments would be adjusted upward each year (increasing by perhaps 1%
annually) until reaching the 12% market rate.
Meanwhile, the difference between actual payments and money owed on
the basis of a 12% interest rate would be added to the outstanding loan
balance. The term of the loan extends until repayment is achieved. This
method is designed to work best for those who can expect their incomes to
grow over time, blowing them either to make larger than required monthly
payments and thus more rapidly reduce the outstanding loan balance, or
to refinance their houses once they can afford a more standard mortgage
With this method 71% of WSHCP participants could afford their houses.
A cautionary note: this option should be used only by those whose incomes
will rise high enough to enable them to make the annually increasing re-
quired payments. Moreover, it must be recognized that if the borrower does
not sell, refinance or, at some point, begin making larger than required
payments, the term of the loan could extend over very long periods of time.
V. Conclusions and Recommendations
The Women's Self-Help Construction Project is an ambitious undertaking
and one which covers difficult terrain. Not only does the project introduce
women into construction activities--an area of work until now the virtual
preserve of men--it also involves them in the controversial area of the
self-help approach to housing solutions. The project raises questions about
the potential impact of skills training on future income generation, the
potentially positive effect of the entire experience on women's lives, and
the chances for project replicability including the issue of cost recovery
for the most impoverished segment of the Panamanian population. Indeed the
project highlights some of the more difficult issues, in both theory and
practice, regarding women's economic roles and the entire concept of self-
What can we conclude from our study of the project? First, and fore-
most, the WSHCP has demonstrated that such projects are worthwhile invest-
ments. The project has produced a stream of tangible and intangible
benefits, providing, in its first phase, 50 families with homes, upgrading
their environment, providing participants with new skills, and generating
self-esteem and community bonds. It is also likely to impact the develop-
ment of succeeding generations. The project's costs, on the other hand,
were not significantly greater than those of similar projects carried out
by the Ministry of Housing (MIVI).
The training provided through the project proved useful on the con-
struction site and enabled the women of the project to build and finish
their houses under the excellent supervision which they received from
SENAFORP. While estimates of the future economic value of training in
construction skills cannot be made until the project is completed and the
participants resume or augment their income-earning activities, it appears
that training for women in non-traditional skills is warranted in terms
of improved community initiative and political returns.11
Nevertheless, it would be worth inquiring in a follow-up study whether
the provision of training in construction is indeed the best use of
SENAFORP's resources to achieve the goal of providing housing for low-
There are indications that replicability of the project may hinge
on the ability to duplicate the extraordinary leadership characteristics
of Romelia Pardo, and the staff of SENAFORP. Yet it is also true, as our
data shows, that the women who participated in the project were typical in
many respects; their backgrounds and experiences were not extraordinary.
Another factor in replicability will be the availability of land for
such projects, and government/social services such as child care and trans-
portation. The provision of transportation to the work site seems to have
been an important factor in the low rates of absenteeism in the WSHCP, and
while women of the project did not use the child care facility provided this
may have been due to their unfamiliarity with the facility, and its loca-
tion which was not on-site. Aside from such services, replicability will
of course require the political will of the government.
As for the difficult issue of cost recovery, our analysis shows that
if conventional pay-back systems are employed on this project and other
similar projects, families with incomes below the 19th decile or those
with a monthly family income below $230 in 1982 prices would be excluded.
That would imply that fully 64% of the WSHCP participants and women like
them, whose families desperately need housing, would be excluded. The
alternative of providing these low-income families with a serviced lot
would provide a partial solution, but one that is not entirely satisfactory
because some women would have to work to build superior housing for their
neighbors while receiving only a foundation for their own families. This
might undermine the community spirit which appears to have sustained the
commitment of the women and ensured the success of the project. In a
choice between serviced lots for the lowest decile or no housing at all,
serviced lots would of course be the preferred solution. Alternatively,
however, more "creative" methods of low-cost housing finance could be con-
sidered. Some of these strategies and options have been outlined in our
section on cost recovery. Several of the more promising cost-recovery
options discussed rely for their success on improvements in individual and
community incomes. Focussing on these options, by developing income-earning
projects within the community, could build upon the community bonds and
development potential which were an integral part of the WSHCP. Unfor-
tunately, given a market interest rate of 12%, no option other than com-
munity payment can provide full cost recovery from households in the lowest
income decile, assuming that households can pay only 25% of monthly incomes
toward housing. Several options require grant elements even to reach above the
lowest decile. On the other hand, it is likely that low-income participants
are accustomed to paying more than one quarter of their income for rental
housing, and would be willing to pay at least as much to own housing, thus
increasing the affordability of the project.
Based on our preliminary findings, which indicate the promising nature
of the WSHCP, it would seem advisable for AID to explore mechanisms for
supporting self-help housing projects similar to the WSHCP. Certainly an
effective first step in such an effort would be to continue to monitor the
progress of the project as it moves into its second phase and reaches com-
pletion. Further, AID could consider the use of "creative financing" op-
tions for cost recovery in such projects, and work on a pilot basis with
self-help housing participants (particularly the women of the WSHCP) to
develop financially viable community-based enterprises--not only to improve
incomes and therefore affordability, but also to enable the skills, disci-
pline, and productive orientation developed in the project to be directed
and to flourish in the future.
Numero de Encuesta
Nombre del encuestador:
Nombre de la encuestada:
Direccion de la encuestada:
Fecha de la encuesta:
Para Uso De La Oficina:
Nombre del revisor: RG/NY
____ I__ __
Nombre de la encuestada
PREGUNTAS QUE SE REFIEREN A LA COMPOSICION DEL GRUPO
1.i Cuantas personas estaban viviendo en su casa durnnte su participation
en el proyecto de construccidn? _
2. (ENCUESTADOR: LLENE EL SIGUIENTE CUADRO CON LA
INFORMATION PARA CADA UNO DE LOS INDIVIDUOS QUE
FORMABAN PARTE DEL GRUPO DOMESTIC DE LA
ENCUESTADA DURANTE LA MAYOR PARTE DEL PERIOD DE
CONSTRUCTION. SI LA ENCUESTADA NO SABE LA EDAD
EXACTA O CUANTO EDUCATION CADA PERSONA TIENE, PIDALE
QUE LE DE UNA APROXIMACION. SI LA ENCUESTADA NO
PUEDE DAR UNA RESPUESTA, ESCRIBA NS (NO SABE) EN EL
CUADRO DE COMPOSICION DEL GRUPO DOMESTIC
M o F)
(Marque S o N)
4 1 I I I~
---- ---- I = I C^
e=mrs de secundaria
4Se mudo Ud. a la
Ciudad de Panamf en
los tItimos 5 aAlos?
(Marque S o N)
(ENCUESTADOR: MARQUE ESPACIO APROPIADO)
3. Tiene Ud. un companero (esposo o marido) que vive en otro lado?
4. iEl contribute dinero en efectivo al ingreso familiar?
5. & Cual es la ocupacion de su marido?
6. < Cual es su ocupacion? Tipo de Actividad?
7. Actualmente, de que trabaja su marido?
8. Y Ud. de que trabaja actualmente?
9. < Cual fu* el trabajo anterior de su marido?
10. d Cual fue su trabajo anterior?
PREGUNTAS QUE SE REFIEREN AL PERIOD ANTES DEL INCENDIO
11. d Cuantas personas yivlan en su casa antes de que ocurriera el
incendio en Curundu? _____
12.i Tenia Ud. trabajo cuando ocurrio el incendio?
13. Si la respuesta es afirmativa,6de que estaba trabajando?
(ENCUESTADOR: ESCRIBA LAS CANTIDADAS APROXIMADAS EN
LOS ESPACIOS CORRESPONDIENTES. MARQUE CLARAMENTE PARA
CADA OPCION SI MENSUAL O SEMANAL. SI LA ENCUESTADA NO
RESPONDED O NO SABE, ESCRIBA NS (NO SABE) EN EL ESPACIO
14. Antes del
incendio, .cua6to era el ingreso mensual o semanal de su
B/. .L5 de su propio trabajo (mensual o semanal);
B/. del trabajo de su esposo e hijos (mensual o
B/. de su negocio (mensual o semanal);
B/. de otras fuentes (mensual o semanal).
15. Durante el ano anterior al incendio, cuanto fue el ingreso familiar
mas alto que obtuvieron, ya sea mensual o semanal?
B/. 'r -- fr o semanal).
16. < Cuanto fue el ingreso familiar mas bajoB/. (mensual o
(ENCUESTADOR: EL VALOR DEL NEGOCIO DEBE SER EL VALOR
ACTUAL, NO EL COSTO PAGADO PAR EL NEGOCIO NI LA CANTIDAD
INVERTIDA EN EL NEGOCIO ORIGINALMENTE.)
17. Antes del incendio,d era Ud. o su familiar propietaria de:
Si Cuantas ~etr
18. d Tenia Ud. algun ahorro antes del incendio?
Noj Si Cudnto B/_,
(ENCUESTADOR: INDIQUE TODAS LAS RESPUESTAS QUE SE
APLIQUEN EN LAS SIGUIENTES PREGUNTAS.)
19, Antes del incendio, Ud. era miembro de cualquiera de las siguientes
organization comunitaria No__ Si
organization de mujeres No' Sil_
cooperative No j Si
grupo religioso No Siy
otro tipo de organization No Si
de auto ayuda
20. Antes del incendio, Ud.habia colaborado
cualquiera de los siguientes
en prestamos de viveres,
servicios o dinero
en una organization femenina
en otros proyectos
con otras participants en
21. Como arreglo su horario diario para poder dedicarle tiempo al
proyecto de construction?
horas menos de sueno por dia
horas menos dedicadas a descansar y divertirse
oras menos en cuidado de ninos
/ oras menos en preparation de comidas
Jhoras menos en trabajo pagado
horas menos en otras actividades, como actividades religiosas
o de la comunidad, visits a familiares y amigos, etc.
horas menos en acarrear agua, comprar viveres, lavar, etc.
(ENCUESTADOR: MARQUE UNA OPCION PARA CADA
22.d Recibio Ud. ayuda de algun miembro de su casa en la construction en
0 = nunca; 1 = a veces; 2 = regularmente; 3 = diariamente
(0) (1) (2)_ (3)
(0)_1) (2) (3)
(0) (1 (2) (3)
(0) 1) (2) (3)
23. Cuando Ud. estaba ocupada con el trabajo de la construccion,d recibio
alguna ayuda de otras personas para
0 = nunca; 1 = a veces; 2 = regularmente; 3 = diariamente.
el cuidado de sus ninos
cocinar y preparar comidas
hacer las compras diaries
lavar la ropa
en forma de dinero
(0) (1)V (2)_
(0) (1) (2)
(0) (1) (2)
(0)' 1)_ (2)_
(0) (1) (2)
(0) (l) (2)_
(ENCUESTADOR: MARQUE SOLO UNA RESPUESTA.)
24. d Quienes fueron los que le ayudaron mas a Ud.?
miembros de su propia casa
otro _(indique quien)
(ENCUESTADOR: MARQUE UNA OPCION PARA CADA CATEGORIA.)
25. ~ Su esposo le ayudo con cualquiera de las siguientes tareas?
0=nunc2; l=a veces; 2=regularmente; 3=diariamente.
cuidado de ni'os
cocinar y preparar comidas
hacer las compras
lavar la ropa
26. d.Sus hijos le ayudaron con cualquiera de las siguentes tareas?
0 = nunca; 1 = a veces; 2 = regularmente; 3 = diariamente.
cuidado de los otros ninos
cocinar y preparar comidas
hacer las compras
lavar la ropa
(0) ()__ (2)_
(0) (1) (2)
(0). 1)_ (2)
(0)_ (1) (2)
"27. Si sus hijos le ayudaron,tuvieron que dejar de ir a la escuela para
(ENCUESTADOR: ESCRIBA LAS CANTIDADES APROXIMADAS EN
LOS ESPACIOS CORRESPONDIENTES. MARQUE CLARAMENTE PARA
CADA OPCION SI MENSUAL O SEMANAL. SI LA ENCUESTADA NO
RESPONDED O NO SABE, ESCRIBA NS EN ESPACIO CORRESPONDIENTE.)
.28. Durante la construction en si, como cuanto ingreso recibio su familiar
de las siguentes fuentes:
B. /c de su propio trabajo (mensual o semanal) L -
Sdel trabajo de su marido e hijos (mensual o semanal)
Sde su negocio (mensual o semanal)
de otras fuentes (mensual o semanal)
29. Pidio dinero tado durante el periodo de construction?
(ENCUESTADOR: SI RESPONDIO QUE SI, PIDALE QUE INDIQUE DE
DONDE RECIBIO DINERO Y CUANTO; SI PRESTO DE MAS DE UNA
FUENTE, LLENE TODOS LOS ESPACIOS CORRESPONDIENTES.
ASEGURESE DE INDICAR LA UNIDAD DE TIEMPO: DIA, SEMANA O
30. Si respondio que si,
de familiares o amigos:
de prestamistas, bancos, casas de prestamos, etc.:
B/. prestadas; B/. a ser pagadas (dfas,
semanas, meses) despu s.
B/_ restadas; 8/. a ser pagadas
semanas, meses) despues.
31. i Participo Ud. en un / (grupo de ahorro y prestamo)?
(ENCUESTADOR: INDIQUE GASTOS SEMANALES.)
32. Durante su participacion en el proyecto, tuvo Ud. gastos relacionados
al trabajo de construction que normalmente Ud. no tiene que hacer?
B/-- para el cuidado de sus ninos (semanal)
B/. para lavanderla (semanal)
B/. _5 para comprar comidas ya preparadas, incluyendo comidas en
B/. |L ppara transport (semanal)
B/. para otros gastos como agua potable, ropa especial o uniforms
para el trabajo (semanal)
(ENCUESTADOR: MARQUE ESPACIO APROPRIADO)
33. Usted era:
de sus companeras de trabajo antes del principio del proyecto.
34. Para poder participar en el proyecto, tuvo Ud. que
mandar a sus hijos a vivir en
casas de otras personas Si No
traer a otra persona que
regularmente no vive con Ud.
a su casa que le ayudara con
las tareas domesticas Si NoV
PREGUNTAS QUE SE REFIEREN A LA CAPACITACION
(ENCUESTADOR: MARQUE SOLO UNA RESPUESTA)
35.4 La capacitacion que Ud. recibio fue util en la construction de las
casas del proyecto? Escoja uno:
no podria haber construido casas sin capacitacion
la capacitacion fue muy until
la capacitacion fue mediamente util
la capacitacion no fue util
.36. Desde que recibio e-I adiestrrmiento, ba buscado
un trabajo en construction
Si No, Por que'no? ,
(ENCUESTADOR: MARQUE SOLO UNA RESPUESTA)
37. Si respondio que si ha buscado trabajo,que tipo de trabajo?
empezo su propio negocio de contratista
sub-contratos con empresas de construction
otro tipo; especifique el tipo
38~ Ha logrado obt er trbao a -,.
construcci6n desde que recibi .el adiestramiento?
No S/ Si
39. La capacitacion que usted recibi le fue util para obtener trabajo?
40. Que problems mayores tuvo Ud. durante su participation en el 11
de dinero No Si '','*t :
de transport (gastos
y/o tiempo necesario) No Si_
con el cuidado de ninos No Si
actitudes negatives o falta
de apoyo: No Si/
de su marido No/ V i
de su familiar No Si
de amistades No Si
de salud (especifique) No_ Si
otros problems, especifique
41.6 Que beneficios le trajo el proyecto a Ud. ?
\^ J^Ji\A^IX/^r/-c C___________________
42.4 Fue alguno de los siguientes de beneficio a Ud.?
mejores perspectives de trabajo
nuevas actividades o servicios
dominion de tecnicas de
albanilerfa o construction
mas cooperation en su hogar
cambios en actitudes en su hogar
43. Ahora que se mude a esta nueva comunidad piensa usted que tendri mas
gastos que antes?
Por que? -' 2-
44. Ha pensado Ud. en obtener un ingreso adicional de alguna otra manera?
45. Qud clase de trabajo le gustaria a Ud. hacer?
46. Ha pensado en trabajar junto con otros participants en este program?
47. Tiene Ud. el adiestramient ecesario para este tipo de trabajo?
48. Qu6 tipo de adiestramiento necesitara Ud. para poder trabajar en este
tipo de actividad?
S49. Qud otros recursos o habilidades necesitard Ud. para poder lograr este
tipo de trabajo?
L4 -%,. /.I ;.