Kerala model

The Kerala model refers to the economic practices developed in India's province of Kerala, which are used to assist in disadvantaged areas.

For more than one reason Kerala is widely known as a special state in India and abroad. Quite often, its demographic characterstics and the achievements on the socio-economic front are highlighted as 'kerala model'(thampi,K.E 2001 Kerala times p12-16) and is unique among developing countries. Kerala,the southern part of India is well known for the world for the first elected ministry of communists. From olden days onwards Kerala women enjoys a leadership role in the society. Kerala has a sex ratio which is more advantageous to women. State has an extensive network of hospitals and primary health centres, and facilities are far better than any developed country.Death rate is far below. Number of women as head of the family is on increase.Participation of the women in politics is almost equivalent to developed countries. This state spending 60% of its annual budget to running public educational network.

'The ration shops' are model for public distribution system of food stuff and is a grand success comparing to any other developed country. The Kerala model is sometimes attributed to Christian missionaries, who lowered illiteracy and weakened the traditional caste system.

Introduction

Delegates to the first International Congress on Kerala Studies in August 1994 focused their attention on shortcomings of and threats to the widely acclaimed "Kerala Model" of development. They were right to do so: the essence of the Kerala Model is social justice, and social justice can only come about through honest discussion of all groups left out of the model's achievements, followed by effective action to bring the "forgotten ones" in.

Kerala has brought near 1st world levels to most of its people in literacy, life expectancy, and infant mortality. Kerala's achievements have largely transcended caste, class, rural/urban and gender limitations (Franke and Chasin 1994). But papers at the International Congress revealed a number of small groups left out. These included fishing people (Karuna et al and Kurien), female stone cutters (Ukkuru et al), female domestic servants (Subramony), some female agricultural laborers (Mencher), at least some tribal peoples (Devi; Corrie), and migrant workers from Tamil Nadu. Although not discussed specifically at the Congress, we could add that headload and other casual laborers may also suffer deprivations not consistent with Kerala's social justice model of development (Pillai 1992).

Adding all these groups together, we estimate that perhaps 15% of Kerala's people are left out in some way from the benefits of the Kerala model. Although most development experts would consider 85% beneficiaries a remarkable achievement, Kerala's activists should see this 15% as a challenge. The Kerala model cannot be fully successful until all the state's people are included.

A closer examination of the "forgotten ones" makes clear that women are a major component. As in other societies, Kerala women earn less than men and face numerous barriers to equality and economic security. Understanding the dimensions of women's inequality is one step towards developing an agenda for action to overcome it.

In this paper, we offer evidence on two dimensions of male/female inequality in the Central Kerala village of Nadur: (1) women's work and wages and (2) the effects of wages on households primarily dependent on women for their incomes. Our research in Nadur took place from 15 November, 1986 to 15 July, 1987. We surveyed 170 households including 1,035 individuals across a range of castes, classes, income levels, family structures, educational levels, and Kerala model variables such as land reform, ration shops, school lunches, and special development programs. Our main village-level findings appear in our study entitled Life Is a Little Better: Redistribution as a Development Strategy in Nadur Village Kerala (Franke 1993), soon to appear in a Malayalam translation. That study covers caste, class, and income inequality, but does not include our findings on women. We present them at this conference for the first time.1

Women's Work and Women's Wages in Nadur

Our sample includes 676 individuals between the working ages of 15 and 64 as shown on table 1. Occupational categories are listed from most to least common. The table shows that unemployment is the main situation for both males and females added together. The first finding confirms what other observers have reported for Kerala: unemployment is the most serious problem facing the work force.

Looking beneath this overwhelming problem, we see that working men and women differ in their employment patterns. "Household affairs" is most common for females while males dominate the category of house compound and general labor with 89 workers against only 23 women. By contrast, women make up 57 of the 64 persons listing agricultural labor as their main occupation.

These first two labor categories are important. In 1987 house compound and general labor brought in Rs 25-30 per day. Agricultural labor garnered Rs 20 for men and Rs 12 for women.

                        Table 1
           Main Occupation of Household Members
                   Ages 15 Through 64
               Nadur Village Kerala, 1986-87
------------------------------------------------------------------
Occupation         Males %   Females  %     Total  Percent
------------------------------------------------------------------
Unemployed          51  15.4   96   27.9   147     21.7
Household Affairs   11   3.3  126   36.6   137     20.3
House Compound and
   General Labor    89  26.8   23    6.7   112     16.6
Agricultural Labor   7   2.1   57   16.6    64      9.5
Student             25   8.4   23    6.7    48      7.1
Petty Trade         30   9.0    3    0.9    33      4.9
Skilled Labor       29   8.7    1    0.3    30      4.4
Service             12   3.6    4    1.2    16      2.4
White Collar        10   3.0    4    1.2    14      2.1
Farmer              11   3.3    0    0.0    11      1.6
Professional         5   1.5    2    0.6     7      1.0
Pensioner            3   0.9    1    0.3     4      0.6
Other                4   1.2    3    0.9     7      1.0

Subtotal           287  86.4  343   99.7   630     93.2

Absent Laborer      45  13.6    1    0.3    46      6.8

Total              332 100.0  344  100.0   676    100.0
------------------------------------------------------------------

Notes: The category Other includes 3 servants, a religious practitioner, and 3 unknown. Spinners at the Cooperative are included in General Labor. Source: adapted from Franke 1993:162.

We see further that women are approximately equal as numbers of students, but table 1 shows that petty trade, skilled labor, white collar, service work, farmer, and professional employment are dominated by men at an average ratio of about 10 to 1. Most Nadur men do not earn high incomes, but nearly all Nadur women are relegated to the status unemployed, household affairs, or the lowest paying agricultural field labor.

Table 1 also shows that 45 of 46 absent workers from Nadur are males. Except for one female working as a maid in New Delhi, all remittance incomes are provided by men. These data are more extreme than Kerala's overall picture, but they fit generally into the Kerala pattern where overwhelmingly workers outside their home villages are men.2 In Nadur in 1987, 9% of all sample income derived from male labor outside the village, further skewing the inequality in earnings between males and females individually within Nadur.

Female-Supported Households as Units of Analysis

Individual employment statistics are important but incomplete indicators of individual well-being. Most individuals in Kerala live in households where income and resources are shared to some extent. Thus, females earning low incomes, suffering unemploy- ment, or engaging in household work, could be receiving some of the benefits of their higher-earning male household partners. Depending on the household structure and composition, these could include husbands (nuclear households), fathers or sons (extended households), male in-laws (joint households), or combi- nations of these (complex households). Even the single female -- widowed, divorced, abandoned, or never married -- could be receiving assistance from male relatives nearby. Especially in rural areas, where family members live close to each other -- even if households have partitioned -- households are usually more appropriate units of analysis than individuals.3

The question then becomes: to what extent does the concentra- tion of females in the lowest income-generating labor categories lower the incomes of the households themselves?

Female-Supported Households in Nadur

We identified 22 female-supported households, or 13% among the 170 households in the 1987 Nadur sample. Female-supported means that over 50% of household income derives from the work of female household members.4 Table 2 shows the distribution of the working members of the two kinds of households.5

                        Table 2
           Main Occupation of Household Members
    Ages 15 Through 64 by Gender of Economic Dependency
               Nadur Village Kerala, 1986-87
------------------------------------------------------------------
                  Male-Supported        Female-Supported
                      N=148                 N=22
                 -----------------       ----------------
Occupation         Males    Females      Males    Females
------------------------------------------------------------------
Unemployed           48        94          3        2
Household Affairs    10       119          1        7
House Compound and
   General Labor     89        14          0        9
Agricultural Labor    6        46          1       11
Student              23        19          2        4
Petty Trade          29         0          1        3
Skilled Labor        29         0          0        1
Service              13         3          0        3
White Collar          7         3          0        1
Farmer               11         0          0        0
Professional          7         1          1        1
Pensioner             3         1          0        0
Other                 3         1          0        0

Subtotal            278       301          9       42

Absent Laborer       42         0          3        1

Total               320       301         12       43
------------------------------------------------------------------

From table 2 we can see that male-dependent households have most of the better paying income sources such as professional, skilled labor, service occupations, white collar, and house compound labor. After subtracting students, unemployed, and household affairs persons, male-supported households have 308 workers or an average of 2.1 per household. Female-supported households, by contrast, have 36 workers for 22 households, or 1.6 on average.

As we predicted, the incomes of the female-supported house- holds were substantially lower than male-supported households; average household income was Rs 3,702 for female-supported compared with Rs 7,342 for male-supported households. This difference was statistically significant at 0.02 in an analysis of variance (ANOVA). As table 3 indicates, female-supported households were distributed far more in the lower income quintiles.

                        Table 3
        Quintile Distributions of Household Incomes of 
     Male-Supported Versus Female Supported Households
                Nadur Village Kerala, 1987
------------------------------------------------------------------
                          Male-Supported  Female-Supported
                         ----------------  ----------------
Quintile                  Number  Percent  Number  Percent
------------------------------------------------------------------
Highest (Rs 9,200-54,000)    33      22        1      5
2nd    (Rs 6,000- 9,199)     33      22        1      5
3rd    (Rs 3,920- 5,999)     29      20        5     23
4th    (Rs 2,900- 3,919)     30      20        4     18
Bottom (Rs   840- 2,899)     23      16       11     50

Totals                      148     100       22    101
------------------------------------------------------------------

Thus we see that 22% of male-supported households are within the best-off quintile compared to only 5% of female-supported households. By contrast, 50% of female-supported and 16% of male-supported households were in the lowest quintile.

General household income, however, does not take into account size differences as we saw above in the breakdown of dependent household members over 64 and under 14. Female-supported households average 3.9 members versus 6.4 for male-supported households. When adult equivalents (AE)6 are used, we find that female-supported households average 3.2 versus 5.4 for male-supported households. When household adult equivalent composition is taken into account, the AE income averages Rs 1,501 for female-supported households versus Rs 1,570 for male- supported ones. The difference is not statistically significant. Similarly, households are almost equally distributed across the income quintiles as shown on table 4.

                        Table 4
     Quintile Distributions of Adult Equivalent Incomes of 
     Male-Supported Versus Female Supported Households
                Nadur Village Kerala, 1987
------------------------------------------------------------------
                        Male-Supported    Female-Supported
                       -----------------   ----------------
Quintile                Number   Percent    Number Percent
------------------------------------------------------------------
Highest (Rs 1,951-9,265)   30       20         4     18
2nd    (Rs 1,252-1,950)    28       19         6     27
3rd    (Rs   857-1,219)    30       20         4     18
4th    (Rs   632-  850)    30       20         4     18
Bottom (Rs   267-  621)    30       20         4     18

Totals                    148       99        22     99
------------------------------------------------------------------

Table 4 and the AE income data for the Nadur sample strongly indicate that households supported primarily by women do almost as well as households supported primarily by males. Since table 1 shows clearly that women are clustered in the lower wage jobs, and the associated wage data show that women earn less than men in comparable jobs such as agricultural labor, we must conclude that female-supported households have developed survival strategies and that these strategies succeed in placing them almost economically equal to male-supported households. What could these strategies be?

Strategy 1: Small Size and Limited Dependents

In the absence of higher-earning male supporters, female-supported households apparently limit their size to maximize their per-AE incomes. To a certain extent this is a consequence of the likely causes of female-supported households in the first place. Most are made up of women whose husbands died, abandoned them, or have become too ill or feeble to work -- along with any dependents these women cannot or are not willing to try to place in other households.

We can see from table 2 that female-supported households absorb less than their share of unemployed, household affairs, and students: with 13% of the households, they account for 3% of the unemployed, 6% of those living off household affairs, and 12% of students. Members over 64 are 15% (13 of 88) in female-supported households along with 7% of children under 14 (21 of 317).7 The demographic support burden on female-supported households is thus less than on their male-supported counterparts. Another way to conceptualize this is to say that female-supported households manage greater efficiency in the relation between earners and dependents.

How do female-supported households achieve this efficiency? One mechanism is to partition themselves off from their children, living in small houses on the family paramba, after transferring any rice land to the adult children. Added to this may be benefits from Kerala government programs such as agricultural laborer pensions.

    Lakshmiyamma is a 71-year-old widow of many years.  
    She lives with her 33 year old daughter Narayani, an 
    agricultural laborer whose husband deserted her 
    several years back.  One of Lakshmiyamma's sons is 
    in a leprosy hospital.  Lakshmiyamma and Narayani 
    live in a small house on the edge of the main para-
    mba.  Two sons live in the main house, managing the 
    1.5 acres rice land to benefit their own partitioned 
    households.  Along with Narayani's small income from 
    field labor, the household lives off Lakshmiyamma's 
    agricultural laborer's pension of Rs 45 per month [in 
    1987].8

Lakshmiyamma is a Nair caste member, like 82% of the female-supported households in Nadur. None are Nambudiri, Muslim, or Mannan; 1 is Pulaya, 1 Chetty, and 2 are Ezhavas. The flexibility of membership and gender roles as heads of households among Nairs may contribute to their being so over represented in the female-supported households. Nairs make up 48% of the Nadur sample households overall.

Lakshmiyamma is one of 10 widows among the 22 female-supported households. Two other widows are receiving agricultural laborer pensions, one receives a pension from a company in Madras where her husband had worked before his death, one works on a rubber estate, another sells coconuts, and another sells rice from the farm land of her children who are living away from Nadur. Another widow of 20 years shares running of a tea shop with two adult daughters; one other widow has a 29 year old daughter who works as an agricultural laborer and another daughter who receives a pension for being handicapped. Finally,

    Janaky is a 65 year old Pulaya widow of 15 years.  
    Three daughters -- 30, 28, and 25 years -- work as 
    agricultural laborers.  One granddaughter of 7 also 
    lives in the household.  The 3 combined agricultural 
    labor incomes are almost all the household has on 
    which to survive.  The daughters sometimes get a 
    little paramba work as well.  In 1986 the household 
    received a Rs 6,000 building loan to repair their 
    house which had been in bad condition.  They are 
    required to repay Rs 4,500 of the loan.

Janaky's case illustrates the efficiency of a low-income household: 3 workers supporting themselves, one elderly mother -- not apparently receiving an agricultural laborer pension to which she is presumably entitled -- and one grandchild. One husband is unaccounted for, and 2 adult daughters remain unmarried -- one of the consequences of their situation.

Altogether 8 males and 54 females are widowers/widows in 59 sample households. Since 10 of these households are female-supported, the chances of becoming female-supported are not increased by widowhood: 17% (10/59) versus 13% of the sample overall (22/170). Similarly, 83% (49/59) of male-supported households have widows or widowers versus 87% of sample overall (148/170). The ratios for each gender of support are thus almost the same.9

With separation, however, the chances are reversed. Six households are headed by women separated, divorced, or deserted by husbands. Four male-supported households have a separated female. The odds of becoming female-supported are 60% (6/10 versus 13% of sample) after separation contrasted with 40% (4/10 versus 87% of sample) for remaining male-supported. In one case, a 68 year old woman lives with her 56 year old sister and the sister's adult daughter. The husband is severely ill, but lives in his mother's family's house in a nearby town. The elderly woman receives a small amount of money for cleaning the main village temple and the sister's daughter brings in a small income as a tailor. Another separated woman of 54 lives by herself and works in a rubber estate; a third now divorced receives rice as needed from her daughter living in the next village; a 46 year old divorced pappadam maker continues to make pappadams while supporting her 23 year old daughter training to be a nurse. One 26 year old woman abandoned by her husband works full time in Nadur's spinning coop from which she earns a meager income. And one divorced woman lives from agricultural labor and work as a servant to a high caste household. Her daughter also does agricultural labor; together they support 2 infant children and the daughter's husband who has been severely ill and weak for several months.

The final 6 female-supported households have intact couples, but are also primarily aged partition split-offs or households with adults unable to work full time because of illness.

    Govindan is a 49 year old Nair former temple servant 
    and agricultural laborer.  Both he and his wife have 
    had health problems and are not able to work regular-
    ly.  The household's 8 members include the parents 
    and their 6 daughters.  Amaru, 19, and Karthyayani, 
    18, work at the Nadur spinning cooperative.  Their 
    combined earnings are the primary support of the 
    household.  When Amaru marries, a prospect soon to 
    happen, she will probably stop cranking the charkas 
    to rewind the thread and the household may have to 
    send the 3rd daughter, Chinnamaru, now 16, and 
    studying in the 10th standard, to help keep the 
    household afloat.10

One Nair household depends entirely on the incomes of 2 spinn- ers to support its 5 members. Another Nair household combines the former village clerk's pension to the 83 year old male head with the income from their 39-year old daughter who works as a teacher in the village school. A daughter-teacher combines with income from temple work to support an elderly couple who lost substantially in the land reform. A household partition combines the agricultural labor pension and servant income of the male head with his adult daughter's work as an agricultural laborer and maidservant to support their small house on the edge of a compound where the larger dwelling has gone over to the other adult children.

Finally, Sankaran and his wife, aged 79 and 62, both receive agricultural labor pensions. They combine these with the income from one daughter who works as an agricultural laborer and another who works as a postal agent to support the four adults and 4 children.

In general, then, we see female-supported households in Nadur maximizing income per person or AE by remaining small and/or by partitioning off from larger family units. We also see in these cases that women maintain small households in part by remaining childless after divorce or separation. Unlike in the Caribbean and some other parts of the world, Kerala women do not apparent- ly gather children around them into matrifocal or female-centred households. The reasons for this pattern are worthy of further study.

Strategy 2: Kerala Redistribution Programs

The examples in section 5 above illustrate another strategy for female-supported households: extensive use of Kerala government programs to redistribute wealth to the poorest households. Although these programs do not target female-supported households in particular, such households seem to benefit out of proportion to their numbers. Three of the programs seem particularly important in Nadur: agricultural laborer pensions, the ration shop, and the Nadur spinning cooperative.

We saw in the cases above that a total of 6 agricultural labor pensions occurred in 5 of the 22 female-supported households (23%). These pensions accounted for 4% of the combined income of these 22 households, compared with 0.6% for the 12 pensions (8%) distributed among the 148 male-supported households.11

Nadur's Ration Shop provides a second government program especially utilized by female-supported households. Female-supported households average 17 cents rice land and 34 cents house compound land versus 33 cents and 88 cents for male-supported households. Neither of these differences is statistically significant, but the rice land difference qualifies female-supported households for greater ration shop access: 77% of rice over the year as compared to 49% for male-supported households.12 We conducted 24-hour recall nutrition surveys in February and July 1987. In February, female-supported households were slightly above, in July slightly below their male-supported counterparts: all averages were within the range of 2,033 and 2,472 calories. According to respondents, female-supported households ate meat only 8 times per year on average. When the all-vegetarian, all male-supported Nambudiri households were removed from the sample, male-supported households averaged 19 times per year to eat meat. Since the ration shop does not supply meat, the meat-eating differences support our interpretation that a substantial portion of the reason for female-supported success in nearly equal calorie intake to male-supported households is their access to and use of the ration shop. Without the ration shop, the female- supported households are at a strong disadvantage; for the foods it supplies, they achieve virtually equality with male-supported households.

Strategy 3: Nadur's Spinning Coop

Nadur houses a unit of the Kerala Khadi and village Industries Association, a spinning cooperative that employed 29 young women from 3 villages in 1987. The women, ages 15 to 33, turn masses of ground and twisted cotton called sliva into refined thread that can be woven on handlooms. Their work -- hand cranking machines called charkas, is called rewinding.

The work is monotonous and unhealthy. The charkas are not electrified.13 The women sit before the machines for hours. Cotton dust fills the poorly lighted and poorly ventilated room. The workers complain of backache, stomach pain, chest pain, and bronchial disorders. Because of the illnesses and because the young women also have substantial household chores, they aver- age only 2.3 weeks work per month, for an average monthly wage of Rs 103.

Despite the physical difficulty of the work, rewinding sliva at the charkas attracts young women from poor households because of the regular work and the relatively high pay. Rs 103 per month for 12 months equals Rs 1,236 per year -- more than many male agricultural laborers could earn in the rice fields in 1987. Twenty of the coop workers came from Nadur, of whom 11 were members of 9 sample households. Spinning wages made up only 1% of total sample income, but averaged 29% for the 9 households with spinners. Three of these households were female-supported: 14% of female-supported households compared with 4% (6/148) of male-supported households. The spinners in the female-supported households worked an average of 2.8 weeks per month, 22% more than in male-supported households. Only one worked at the average while all the others worked above it. They contributed 73%, 85%, and 100% to their household incomes. Two of the households had 2 spinners each, while:

    Sujatha was married at 23, but her husband deserted 
    her a few months later.  Now 26, she lives alone in a 
    small house on a small plot of paramba.  She has no 
    rice land and her immediate relatives shun her, 
    blaming her for the loss of her mate.  He lives in a 
    nearby village, married to another young woman, and 
    sends no money.  Without her full-time coop job, 
    Sujatha would have only a few coconuts on which to 
    live.

Female-Supported Households and the Kerala Model:

A Continuing Agenda?

The Nadur household survey indicates that female-supported households achieve near equality with their male counterparts. They achieve this near equality despite women's lower wages and their concentration in the lowest paying jobs. The strategies they use include tailoring their household structure to maximize income per consumption unit, making extensive use of Kerala's redistribution programs, and working at unhealthy, unpleasant jobs to earn incomes. As Kerala's planners and activists consider initiatives for the next generation of the Kerala model, each of these strategies raises issues that suggest the need for further research or for immediate political action.

Structure

Let us consider first the structure of female-supported house-holds. While highly efficient in a technical sense, they may be exacting a price in exhaustion from their members. By maximizing workers in the household, female-supported households may lack adequate personnel to take care of household tasks. Or, the members may have to do these tasks after hard days' of labor. In other words, female-supported households may achieve income equality with male counterparts at the cost of concentrating the double-shift within their household boundaries.

To get an indication of the access to household labor, we compared all persons of all ages listing household affairs as their main activity by gender of main support. The 22 female-supported households have 17 household workers (8 between the ages of 14 and 64 -- see table 2) for an average of 0.8 per household. The 148 male-supported households have 183 house-hold workers (129 between 14 and 64) for an average of 1.2, 50% more household workers. In addition, male-supported household affairs workers average 52.4 years compared to 61.4 in female-supported households.14 Female-supported households have fewer workers taking care of the home, and those workers are older and less physically able.

These observations leave several questions unanswered. Do female-supported households that are partitions farm out house-hold tasks to their nearby offspring? Do women workrs in female-supported households make extra use of Kerala's extensive tea shops for meals? Most importantly, perhaps, don't female-supported households have fewer household tasks? We saw on table 3 that most unemployed dependents are in male-supported households. And in section 5 we saw that only 7% of children live in the female-supported households. Therefore, such households have fewer household maintenance burdens. However, all households have certain basic "start-up" activities and daily maintenance activities that require some input from adult household workers: shopping for food, cooking, cleaning, and child care where relevant.

Does the lower demand for household labor balance out for female-supported households or do the adult members suffer more physically and mentally from their double shift burdens? This seems an important area for further research. One approach would be time allocation studies comparing male- and female-supported households to find out how work is carried out inside and outside the households. This could be combined with attitude surveys and/or life histories to get some indications about how the household members feel about their situation.

Life histories might also make possible some understanding of another aspect of female-supported households: the apparent lack of sexual and child-bearing fulfillment available to their adult female members. In India, separation and abandonment leave men in position to have new spouses and families; for women this may be a difficult burden to overcome. Sujatha's case in section 7 above is illustrative. Knowing more about the chances for remarriage for abandoned women in Kerala would help assess the emotional price paid for the break-up of the household that is one major cause for the creation of female-supported households. How many of the cases were freely chosen by the women?

Widowhood poses a similar problem. One estimate for India as a whole gives 33% of ever-widowed women as remarried compared with 66% of men (Gulati 1992:WS95). The 1961 Indian census suggests that 20% of rural Kerala female widows remarry, with no figure for males (Gulati 1992:WS96).

Kerala Redistribution Programs

The Nadur data indicate strongly that 2 of Kerala's major redistribution programs are crucial to female-supported households. Recent central government policies to make the ration shops more like the open market will likely harm female-supported households. Our data suggest that returning the ration shop policies to their previous benefit patterns and maintaining the agricultural labor pensions should be high priorities for activists of Kerala's next phase of development.

Creating Female-Worker-Friendly Rural Employment

Government programs such as ration shop price subsidies and pensions can only be secondary measures in Kerala's development. One of the highest priorities might be rural employment. We saw in section 2 and on table 1 that Kerala women do not go out of their home villages to find work nearly as easily as do men. One solution to this problem is to bring employment to the villages. Nadur's Khadi Cooperative which we describe in section 7 provides some information on the dimensions of this issue. While offering work crucial to the support of several households, the coop until recently took a heavy toll on its workers' health and happiness. In 1992, the coop built an improved structure and electrified some of the machines, but cotton dust still fills the air and workers sit in uncomfortable positions without back rests. The coop building also lacks other amenities such as a convenient, dust-free area for eating lunch.

One way to think about the future of rural employment in Kerala is to look at Kerala Dinesh Beedi Workers Cooperative Society (KDB). This large centralized and decentralized industri- al concern in Kannur and surroundings has managed to pay high wages and maintain profits while introducing gradual improvements in working conditions and benefits. Over 50% of work sheds are now modernized and a building program is underway to modernize the rest within a few years.

Significant features of KDB are its structure and its ownership system. The structure includes centralized purchase and marketing combined with decentralized production units close to the homes of workers. KDB is essentially worker owned. With more than 50% of its workers female by 1992, KDB has become probably the largest female-worker-owned company in the world. Close attention to the details of KDB's success and to its continuing problems might suggest how organizations like Nadur's Khadi cooperative could be modified to make them more attractive to rural workers.15 Such actions seem consistent with the ideals behind Kerala's New Democratic Initiatives that call for energizing the working population for action in its own self-interest. The alternative is to await foreign investment or other outside interventions that may be based on concepts of efficiency intended to produce greater profits through exploitation and abuse of workers rather than through worker participation and social justice in the work place.

Conclusions

Social movements in Kerala historically have engendered hope among workers and farmers that life can be a little better. The strength of these movements has also stimulated optimism and idealism among Kerala's intellectuals and academics. Both groups are represented at this conference on Women in Kerala: Past and Present. The old Kerala model has provided Kerala with many of the elements needed to plan and carry out a new Kerala model. We hope that the needs of women workers and female-supported households will be directly addressed by research and action coming out of this seminar.

Notes

1. A detailed description of Nadur Village and its environment appears in chapter 3 of the English edition of Franke 1993. The rationale for choosing the village is given in chapter 2.

2. Thomas Isaac (1992:4; 1993:64) provides evidence that about 14% of Kerala workers in Kuwait in 1990 were maids, with nearly all other categories likely to be male workers except for teachers, cooks, clerks, and paramedics, of whom a small percentage might be female. Oberai, Prasad, and Sardana (1989:28) found that 15% of Kerala migrants within India were female, a percentage higher than from Bihar (9%) or Uttar Pradesh (13%), the other two states represented in their sample.

3. See Franke 1993:32-33 for further discussion of the rationale for using households rather than individuals for income analysis. Of course, individuals must be the units when intra-household variations or inequality are the subjects of research. For male-female inequality, this is an important area in which much additional research is needed.

4. We chose female-supported over female-headed. For economic analysis, the gender of economic support seems more straightforward. The category female-headed implies concepts of authority as well as income generation. In our Nadur sample 61 respondents (36%) listed a female head of household. In many cases, this seemed to be more a sign of respect towards an aging woman than an accurate account of decision-making power. Some researchers in India would argue that any household with a male 18 years or older could not effectively have a female head. Mencher (n.d.) gives various examples of the dimensions of this problem. She too, seems to conclude that gender of support is the more useful concept.

5. Marilyn Cohen's (1992:309) findings from the 1901 Irish census demonstrate striking similarities with Nadur in patterns of female-supported household structure and composition as well as employment efficiency (1992:306 and 308). This suggests that in male-dominated societies, female-supported households may have similar properties and similar causes despite otherwise large cultural differences.

6. An adult equivalent designates the amount of food energy required to sustain "an average adult male doing sedentrary work." The Indian Council of Medical Research adds tenths of an equivalent for heavier work and subtracts tenths for smaller body size of females and children giving a range of 0.4 to 1.6. Since most Nadur households spend well over 75% of their income on food, we consider the adult equivalent more accurate than per capita. See Franke 1993, chapters 2 and 10 for further details on the use of this measure for the Nadur sample.

7. Males 60+ were 5.9% of sample individuals compared with 7.1% for the same category for Kerala as a whole. Nadur females 60+ were 6.5% compared with 8.6% for all-Kerala. The ratio of females to males 60+ was 1.1 for both (all-Kerala figures for 1986 from Gulati 1992:WS95). These data confirm our claims (Franke 1993:chapter 2) that the Nadur sample is representative. Gulati's calculation of the totals is an average of the two percents rather than their addition, however.

8. The village name Nadur and the names of all individuals are pseudonyms. No other information has been altered.

9. For Kerala in 1981, 19% of males over 70 remained widows. For Nadur, the figure was 21% (6 of 29). Kerala females over 70 were 81% still widowed -- that is, not remarried, compared to 73% for the Nadur sample (19 of 26). For all-Kerala in 1981, the expected duration of widowhood for men was 3.5 years; for women it was 14.7 years (Gulati 1992:WS97 and WS99). We do not have a comparable figure for Nadur, but the combination of longer female life expectation and lower ability to remarry could mean for Nadur and for Kerala that elderly female-supported households are likely to continue to be produced.

10. See Franke 1993:188-190 for a more detailed description of this household.

11. A Chi-Square test on the numbers of households receiving or not receiving at least one pension by gender of main support is significant at p < .05.

12. The ration shop percentage difference is significant in ANOVA at p < .001. For this and all food and nutrition-related compari- sons, we removed the single member households of both genders as consistent with the approach adopted in Franke 1993 and 1993a, where explanations for the decision are provided. On the individual nutrition surveys, ration shop usage by gender of household support was different in the same direction, but not statistically significant.

13. In 1992 the coop built a new work shed with electrical hook- ups for some machines.

14. The age difference is significant at p = 0.0425 using ANOVA.

15. A forthcoming book on KDB will hopefully provide some of the material for the relevant discussions. See Thomas Isaac, Pyaralal Raghavan, and Richard W. Franke, Mobilization, Skill, and Social Justice: Achievements and Dilemmas of Kerala Dinesh Beedi, to be published in Malayalam and English in 1995.

References

Chasin, Barbara H. 1990. Land reform and women's work in a Kerala village. East Lansing. Michigan State University Women in International Development. Working Paper #207.

Cohen, Marilyn. 1992. Survival strategies in female-headed households: linen workers in Tullylish, County Down, 1901. Journal of Family History 17(3):303-318.

Corrie, Bruce P. 1994. The Kerala model of development from the perspective of the dalit child in Kerala. Paper presented at the First International Congress on Kerala Studies. Thiruvananthapuram, 27-29 August 1994.

Devi, T. 1994. Sexual abuse and poverty: tribal women of Wayanad. Paper presented at the First International Congress on Kerala Studies. Thiruvananthapuram, 27-29 August 1994.

Franke, Richard W. 1993. Life Is a Little Better: Redistribution as a Development Strategy in Nadur Village Kerala. Boulder, Colorado. Westview Press. Studies in Conflict and Social Change.

_____. 1993a. Feeding programs and food intake in a Kerala village. Economic and Political Weekly 27(8/9):355-361.

Franke, Richard W. and Barbara H. Chasin. 1994. Kerala: Radical Reform as Development in an Indian State. Oakland, California. Food First. 2nd edition. New Delhi. Promilla and Co. Publishers, Inc. 2nd edition.

Gulati, Leela. 1992. Dimensions of female aging and widowhood: insights from Kerala experience. Economic and Political Weekly 27(43-44):WS93-WS99.

Karuna, M. S. 1994. Socioeconomic status of fishermen families in Thiruvananthapuram District. Paper presented at the First International Congress on Kerala Studies. Thiruvananthapuram, 27-29 August 1994.

Kumar, Rachel. 1994. Development and women's work in Kerala: interactions and paradoxes. Economic and Political Weekly 29(51-52):3249-3254.

Mencher, Joan P. 1986. Female-supported/female-headed households in India: who are they, what are they, and how do they survive? Paper presented at the American Anthropological Association. December 1986.

_____. n.d. Survival strategies among female-supported households in South India. Unpublished manuscript.

_____. n.d. Women's work and poverty: women's contribution to household maintenance in two regions of South India. Unpublished manuscript.

_____. n.d. Women agricultural labourers and land owners in Kerala and Tamil Nadu: some questions about gender and autonomy in the household. Unpublished manuscript.

Mencher, Joan P. 1994. The Kerala model of development: the excluded ones. Paper presented at the First International Congress on Kerala Studies. Thiruvananthapuram, 27-29 August 1994.

Oberai, A. S.; Prasad, Pradhan H.; and Sardana, M. G. 1989. Determinants and Consequences of Internal Migration in India. Delhi. Oxford University Press.

Pillai, S. Mohanan. 1992. Social Security Schemes for Workers in the Unorganised Sector: A Case Study of Headload Workers Welfare Scheme in Kerala. Dissertation. Master of Philosophy in Applied Economics. Trivandrum. Centre for Development Studies.

Subramony, Dhanalakshmy. 1994. A profile of urban poverty: the case of domestic servants in Thiruvananthapuram City. Paper presented at the First International Congress on Kerala Studies. Thiruvananthapuram, 27-29 August 1994.

Thomas Isaac, T. M. 1992. Economic consequences of the Gulf crisis: a study of India with special reference to Kerala. New Delhi. International Labour Organisation Asian Regional Team for Employment Promotion (ARTEP).

_____. 1993. Economic consequences of the Gulf crisis: a study of India with special reference to Kerala. In Wickramasekara, Piyasiri, ed. The Gulf Crisis and South Asia: Studies on the Economic Impact. New Delhi. ARTEP. Pp. 59-102.

Ukkuru; Mary P.; Jyothi Augustine; Prema L.; and Sujatha A. 1994. Working pattern and nutritional profiles of women engaged in stone breaking. Paper presented at the First International Congress on Kerala Studies. Thiruvananthapuram, 27-29 August 1994.