INFLUENCE OF FEMALE AND MALE INCOMES ON FOOD

EXPENDITURES OF NICARAGUAN HOUSEHOLDS

INFLUENCIA DE LOS INGRESOS FEMENINOS Y MASCULINOS SOBRE

EL GASTO EN ALIMENTOS DE LOS HOGARES NICARAGÜENSES

Gómez-Valle, Roxana1

__________________________________________________________________________________________________________________________

Autor por correspondencia: gomezvrs@gmail.com (Gómez-Valle, Roxana)

Forma sugerida de citación: Gómez-Valle, R. (2023). Influence of female and male incomes on food expenditures of Nicaraguan

households. Apuntes de Economía y Sociedad, 4(1), pp. 94-104. https://doi.org/10.5377/aes.v4i1.16441

Conflicto de intereses: La autora ha declarado que no existen ningún conflicto de interés.

Resumen

Otras investigaciones han explorado si los ingresos del

hogar son utilizados de la misma manera dependiendo del

género de la persona que los genere. Sin embargo, la

mayoría de estos estudios no incluyen dentro del análisis a

los hogares con múltiples miembros generadores de

ingreso. Este trabajo intenta llenar este vacío en la literatura

explorando los cambios en los gastos alimentarios de los

hogares generados ante una variación en los ingresos de las

mujeres frente a aquellos percibidos por los hombres, tanto

en los hogares nucleares con dos generadores de ingresos

como en los formados por múltiples miembros generadores

de ingresos. Este estudio es exploratorio y utiliza la

Encuesta de Medición del Nivel de Vida (EMNV) de

Nicaragua de 2014 como fuente de información. Para ello,

se estima una curva de Engel de alimentos, utilizando

análisis de regresión, que integra la noción de coaliciones

de intereses de género como parte de las variables

demográficas. No se encontraron grandes diferencias entre

las asignaciones del presupuesto del hogar destinado a

alimentación frente a los cambios de los ingresos laborales

generados por mujeres y hombres. Las características

personales analizadas afectan de forma negativa el gasto en

alimentación. Por el contrario, las características del hogar,

i.e., número de personas generadoras de ingresos y área de

residencia, son los principales determinantes del aumento

de los gastos alimentarios del hogar. No se encontraron

diferencias entre hogares nucleares y múltiples generadores

de ingresos.

Palabras clave: ingresos, género, gastos en alimentos, curva

de Engel, Nicaragua.

Abstract

Many studies have explored whether household income is

used in the same way depending on the gender of the person

who earns it. However, most of these studies do not include

in the analysis multiearner households. This paper attempts

to fill this gap in the literature by exploring changes in

household food expenditures generated because of a

variation in the income earned by women compared to the

one earned by men, both in dual-earner nuclear households

and in multiearner households. This is an exploratory study

and uses the 2014 Nicaragua Living Standard

Measurement Study (LSMS) as the source of information.

For this purpose, a Food Engel Curve is estimated, using

regression analysis, which integrates the notion of gender

coalitions of interest as part of the demographic variables.

No large differences were found between household food

budget allocations when comparing the changes in labour

income earned by women and men. The personal

characteristics analysed, negatively affect the food

expenditures. In contrast, household characteristics, i.e.,

number of income-earning persons and area of residence,

are the main determinants of the increase in household food

expenditure. No differences were found between nuclear

households and multiearner households.

Keywords: incomes, gender, food expenditures, Engel Curve,

Nicaragua.

1Investigadora Independiente, Nicaragua, Email: gomezvrs@gmail.com ORCID: https://orcid.org/0000-0001-8087-2148

DOI: https://doi.org/10.5377/aes.v4i1.16441

Recibido 23/03/22 – Aceptado 05/06/23

Apuntes de Economía y Sociedad N.º 4 (1) junio 2023/ 94-104/ ISSN: 2709 - 7005

I- Introduction

The reasoning for targeting women as recipients of economic resources in public policy lies in the assumption that women's

contribution, access, and control over resources improve household well-being. Incomes have been of particular interest to

empirically test this assumption. Many studies (Armand, Attanasio, Carneiro & Lechene, 2016; Bhupal & Sam, 2014; Qian,

2008; Salas Alfaro & Baca Tavira, 2015) have been conducted to answer whether household outcomes, including household

expenditure patterns, differ depending on the identity of the earners, e.g., whether incomes are brought by male or female

household members, with most of them favouring the hypothesis and providing evidence against the unitary models of the

household.

A large part of the literature relating individual income contributions with household outcomes has been focused on low and

middle-income countries, but only a few (Angelucci & Garlick, 2014; Djebbari, 2005; Gummerson & Schneider, 2013;

Kazianga & Wahhaj, 2016) have integrated the complexity of household structures of these countries into the analysis. Latin

American households are characterized by having complex arrangements where multiearner households are as common as

dual-earner nuclear households (Esteve, Castro-Martín, & Castro Torres, 2022; Esteve & Florez-Paredes, 2018). In Nicaragua,

the proportion of non-nuclear households adds up to 63.04 per cent (INIDE, 2016a). Besides the intricacies of household

composition in the country, the progress made over women's participation in the labour market has not been without polarities.

The labour rate participation for women has increased over the last years from 40 per cent in 2000 to 49 per cent in 2010 and

53 per cent in 2019 (World Bank, nd), the gender wage gap has closed moderately (Herrera, Dijkstra, & Ruben, 2019), and

female-maintained households, i.e., households where female incomes constitute the largest proportion of household incomes,

represent 35.76 per cent of Nicaraguan households (INIDE, 2016a). At the same time, gender roles are still a strong

determinant of women’s participation in the labour market, e.g., women inserting in low-paid economic sectors (Herrera et

al., 2019), and of perceptions of women’s incomes contributions to the household (Bradshaw, 2013), i.e., women’s incomes

perceived as supplementary to men’s incomes. While being one of the poorest countries in Latin American and the Caribbean,

the general poverty rates in the country (consumption-based) have decreased over the last years from 47.9 per cent in 1998 to

29.6 per cent in 2014 (INIDE, 2016b). This contextual background makes Nicaragua an interesting case of study to see

whether female-generated incomes are used differently than male-generated incomes for the demand for food and whether

these differences occur for any household composition.

This study follows an exploratory approach to uncover any differences in the share of household food expenditures, as a

household outcome, as a result of changes in income sources. More specifically, it answers the research question: how does

the demand for food change with a change in male vis-à-vis a change in female-generated income? Alternatively, are female-

earned labour incomes spent equally on food as male-earned labour incomes? It also explores differences in food expenditures

between nuclear dual-earner households and multiearner households. These questions are answered by estimating an empirical

food Engel curve on its quadratic generalization of the Working-Lesser specification (Lesser, 1963) using data from the 2014

Living Standard Measurement Study (LSMS) for Nicaragua,

co-managed by the World Bank and the Instituto Nacional de

Información de Desarrollo (INIDE) (Zuniga Gonzalez, 2010). This paper does not attempt to test any of the economic models

of household behaviour nor try to establish causal relationships between the measures of interest, but it provides a first

exploration of the intrahousehold distribution of income using regression analysis.

The paper is structured as follows: The theoretical framework guiding the research is presented after the introduction. Section

3 describes the methodology used for the analysis and its limitations. The results are presented and discussed afterwards.

Section 5 concludes.

II- Theoretical framework

2.1. Models of household behaviour and household expenditure patterns

In the last decades, many theoretical economic models have been developed to explain whether social or economic relations

within the household affect observed economic outcomes. The unitary model assumes that households have one utility

function and one primary decision-maker who altruistically allocates the household resources to maximize utility. Under this

model, household members have shared preferences and the same marginal utility of income (Elsas, 2013). Following the

unitary model, the amount and composition of household expenditures depend only on the pooled incomes but not on who

generates them (Abdullah Yusof & Duasa, 2010; Browning, Bourguignon, Chiappori, & Lechene, 1994).

Apuntes de Economía y Sociedad N.º 4 (1) junio 2023/ 94-104/ ISSN: 2709 - 7005

Whether the household behaves as a unit or not may influence the results of policy interventions. Quisumbing (2003) identifies

four types of policy failures in case the unitary model were rejected: i) not obtaining desired outcomes of public transfers if

some recipients allocate the resources more "efficiently" than others; ii) if resources, including information, are not shared, a

wrong targeting compromises the entire intervention; iii) having an imperfect/incomplete policy design, in case factors

different than prices, incomes, or exogenous endowments impact the allocative outcomes; and iv) the dilution of transfers'

objectives if some household members adjust their behaviour differently to new sources of income.

Consequently, a large body of empirical work has been oriented to test the unitary model (Bircan Bodur & Mukiyen Avci,

2016; Doss, 2013), mostly by comparing whether (and how) household expenditures change given different income sources,

obtaining mixed results (Opata, Ezeibe & Ume, 2020; Phipps & Burton, 1998; Thomas, 1993). These results discrepancies

gave birth to the conventional collective models - cooperative and non-cooperative models (Alderman, Chiappori, Haddad,

Hoddinott & Kanbur, 1995). Collective models allow the possibility of divergent preferences between household members

and recognize the role of individual economic resources to maximize individual desires and impose preferences. Under

collective models, the source of income will influence the observed outcomes, including expenditure composition.

A large part of the literature studying the relationship between incomes and expenditure composition in Latin America are

impact assessments of the PROGRESA/OPORTUNIDADES/PROSPERA (POP) program, a conditional cash transfer (CCT)

intervention directed to improve welfare conditions in rural Mexico. Bobonis (2009) and Attanasio and Lechene (2002), find

out that female incomes positively affect public goods expenditures more than those from alternative sources, especially on

children's clothing. Similarly, Attanasio and Lechene (2010, 2014) find that the impact of the transfers to women on food

consumption differs from the result expected by the unitary model and instead follows the patterns of an efficient collective

model. Other studies assessing the impact of CCT targeting women as recipients on the share of food expenditures in Colombia

(Attanasio, Battistin & Mesnard, 2011) and Nicaragua (Maluccio & Flores, 2004) and from an unconditional cash transfer

(UCT) program in Ecuador (Schady & Rosero, 2008) have obtained similar results.

2.2. Multiearner households

The evidence on expenditure compositions for Latin America has usually used nuclear households as the unit of study, yet

household composition in the region is more complicated. Outcomes, resource allocations, and decision-making processes

may vary according to family types (De Backer, Holvoet & Milanzi, 2021; Djebbari, 2005; Hayashi, 1995; Kazianga &

Wahhaj, 2016), whence the comparison of outcomes for different household arrangements become necessary. Even nuclear

households may have different age compositions, resulting in members generating incomes beyond the married dyad. Phipps

and Burton (1998) recognize that adult sons and daughters generating income may have some space to decide how they should

be spent. Shared preferences for parents and non-dependent children are less likely to occur because of age and hierarchical

relationships. Parents may try to maximize their children's welfare, but it is less likely that it occurs, e.g., between siblings.

Divergent preferences in multiearner households may be simplified by the emergence of interest coalitions. It means that

individuals with similar preferences create groups to position themselves in the allocative process (Gummerson & Schneider,

2013; Pickbourn, 2016). It also opens the possibility of resource sharing within the coalitions. Gender has been pointed out

as a crucial factor in shaping member preferences. Schmeer (2005) emphasizes that "[…] gender norms and stratification

systems create differences in women's and men's roles and responsibilities, resulting in husbands and wives having distinct

interests, power, and spending preferences" (p.400). This statement may be extended to different members having similar

preferences according to their position in social life.

Nevertheless, the analysis needs to be careful when preferences over outcomes are not observed or revealed. For these events,

preferences are usually assumed based on groups, especially gender (Doss, 2013). Thomas (1990) calls this an "inferential

approach" for preferences, i.e., the inference that a person does prefer an observable outcome.

For the Nicaraguan case, not empirical evidence was found exploring the relationship between incomes and household

expenditures that consider the role of multiearner households in the allocative process. The rest of this paper explores how

the household demand for food changes with a variation in male incomes vis-à-vis a change in female incomes, with the

hypothesis that the identity of the income providers should not generate any changes in the demand for food. More specifically,

it explores how female and male-generated incomes are related to changes in the household demand for food, being aware of

any difference between nuclear dual-earner and multiearner households. It is worth mentioning that this study does not pretend

to do formal tests of the unitary model but to advance an overview of the relationship between income sources and allocative

outcomes in contexts where the nuclear household is not representative of the household structure.

Apuntes de Economía y Sociedad N.º 4 (1) junio 2023/ 94-104/ ISSN: 2709 - 7005

III- Methodology

3.1. Empirical approach

This paper follows a quantitative and exploratory approach and uses regression analysis to establish the non-causal

relationship between female and male-earned incomes and household food expenditures. This relationship is estimated using

an empirical food Engel curve on its quadratic generalization of the Working-Lesser specification (Lesser, 1963). The function

to be estimated is represented by:

   

  

     󰇛󰇜

where  is the household share of food,  represents the natural logarithm of female-earned household incomes and 

 its

quadratic form,  designates the natural logarithm of male-earned household incomes, also included in its quadratic form



;  represents the aggregated prices for food;  is a vector of demographic characteristics attributable to the household

members, following the interest coalition assumptions, and  is a set of household variables;  and  are the stochastic

disturbance and constant of the model, respectively.

Equation (1) is estimated using cross-sectional data from the 2014 Nicaragua LSMS (INIDE, 2016a). The 2014 Nicaragua

LSMS is the last openly available nationally representative survey that collects data on household expenditures, and individual

earned incomes. The survey has a sample of 6,851 households. However, only those households with at least one female and

one male earner are included in the analysis leading to 2,986 households and 2,981 observations without missing values.

Figure 1 describes the households included in the analysis. The number of observations per variable, and other descriptive

statistics, are presented in Table 1.

The dependent variable, the household share of food, is estimated as the percentage of the monthly household expenditures

assigned to food. Household expenditures for food include cash expenses, the value of goods produced for self-consumption

and those obtained from small businesses owned by the household members,

and the demand for prepared meals consumed

inside the household. Expenditures in prepared meals consumed outside the household were not included under this category

since the survey did not provide additional information to identify whether these were demanded as a leisure activity, e.g.,

meals in restaurants under the context of social events or as food away from home consumption. The LSMS collects

information of food demanded over the last 15 days.

Figure 1. Flowchart of households included in the analysis.

Apuntes de Economía y Sociedad N.º 4 (1) junio 2023/ 94-104/ ISSN: 2709 - 7005

The price for food is approximated as the sum of the prices of each good consumed by the household in the local currency,

córdoba oro (CO). For those goods that were obtained from the market and produced by the household members, the price

was estimated as the average between the market and the estimated value to account for good differentiation. The analysis did

not include expenditure categories different from food, given the lack of information about their market prices.

The 2014 Nicaragua LSMS collects information on the earned labour incomes of all household members above ten years old

at the individual level. For dual-earner nuclear households, each of the couple's incomes constitutes household female-

generated and male-generated incomes. For multiearner households, i.e., a household with multiple adult earners, the paper

follows the construction proposed by Gummerson and Schneider (2013) under the assumption of gender interest coalitions.

For these households, female-generated incomes represent the total amount of earned labour income generated by the female

household members; analogously, male-generated incomes are the addition of all earned labour incomes generated by the

male household members. Multiearner households are not only constituted by nuclear households with adult children earning

an income but any other household configuration with at least three earners, e.g., three-generation households.

Similar to the construction of incomes, the variables reflecting personal demographic characteristics were aggregated by

gender group. The variables include the average age and education of female and male earners, as suggested by previous

research (Gummerson & Schneider, 2013). The literature (Doss, 2001, 2006; Schmeer, 2005; Mulamba, 2022; Tingum &

Kuponiyi, 2020) also guided the selection of correlates indicating household characteristics, namely the number of adult

earners in the household (female and male), the natural logarithm of the household size,

iii

the number of children in the

household, the area and region of residence and the household structure (dual-earner nuclear and multiearner households).

Table 1 presents the variables used with their descriptive statistics.

Table 1

Summary statistics

Variable

Observati

ons

Mean

Std.

Dev.

Min

Max

Notes

Share of food

expenditures (%)

2,981

44.52

14.30

0.64

88.73

Food expenditures over the last 15 days

in monthly frequency.

Percentage over the total household

expenditures.

Female incomes (ln)

2,986

8.26

1.11

1.67

12.61

In the local currency (CO) and monthly

frequency.

Labour incomes derived from first and

second job of the past week and work of

the last 12 months.

Female incomes (squared)

2,986

69.52

17.77

2.77

159.05

Male incomes (ln)

2,986

8.79

1.03

3.54

12.38

In the local currency (CO) and monthly

frequency.

Labour incomes derived from first and

second job of the past week and work of

the last 12 months.

Male incomes (squared)

2,986

78.32

17.28

12.50

153.38

Food price (CO)

2,981

500.30

241.29

30.00

2,605.27

In the local currency (CO) and monthly

frequency.

Sum of prices for each good demanded.

Average age of female

earners (years)

2,986

30.22

12.28

7.50

79.00

Only those earning an income included

Average age of male

earners (years)

2,986

28.97

12.52

7.00

90.00

Only those earning an income included

Years of education of

female earners (years)

2,986

6.78

3.86

0.00

20.00

Only those earning an income included

Years of education of

male earners (years)

2,986

6.56

3.91

0.00

21.00

Only those earning an income included

Apuntes de Economía y Sociedad N.º 4 (1) junio 2023/ 94-104/ ISSN: 2709 - 7005

Number of female earners

(linear)

2,986

2.58

1.45

1.00

14.00

Only those earning an income included

Number of male earners

(linear)

2,986

2.44

1.34

1.00

10.00

Only those earning an income included

Housheold size (ln)

2,986

1.53

0.41

0.69

3.04

Derived from the number of household

members

Area of residence (binary,

rural)

2,986

0.29

0.45

Rural housheolds used as the base

Region of residence

(macroregions)

Managua (capital) used as the base

Rest of Pacific

2,986

0.27

0.44

Central region

2,986

0.31

0.46

Caribbean region

2,986

0.12

0.33

Number of children

(linear)

2,986

1.38

1.26

0.00

10.00

Under 14 years old

Household structure

(binary, nuclear)

2,986

0.05

0.22

Nuclear households with or without

dependant children

Author's elaboration based on LSMS (2016a). All results are weighted.

Variables with high variance were checked for high leverage data points (atypical values). Influential leverage, presumably

not derived from data entry errors or related, was found for one of the independent variables (effect on its regression

coefficient). Given that the estimation using Ordinary Least Squares (OLS) was too sensitive to these values, a robust approach

is used to estimate equation (1). Since “bad leverage points” were found in one independent variable an estimator with a high

breakdown point that could deal with this type of outlier was selected based on Verardi and Croux (2009). A robust MM-

estimator is used in this study. Given the small number of dummy variables among the regressors, these are not expected to

affect the estimator. It is worth mentioning that robust regression analysis is not without its critics (Brossat, Parker & Castillo,

2011), but not reasons of concern were found for this analysis. Moreover, robust standard errors were used to “correct” for

heteroskedasticity in the dependent variable. Both, the selection process for the estimator and heteroskedasticity tests can be

followed with the data shared with this paper.

3.2. Limitations in the methodology

Given that the LSMS 2014 does not contain information on unearned incomes, e.g., remittances, at the individual level, the

analysis is made using earned labour incomes only, which may underestimate the availability of resources of each gender

interest coalition group and the final effect on food expenditures (Armand et al., 2016). More importantly, the 2014 Nicaragua

LSMS is designed to estimate living standards; therefore, it has no specific information about household members'

preferences or control over earned incomes. The results in this study should be read as the use of earned labour incomes

coming from different sources, i.e., male and female-generated incomes, and not as female or male control or decision-making

over incomes.

IV- Results

Table 2 presents the results from the regression analysis. The results support the hypothesis that female-generated labour

incomes lead to higher shares of household food expenditures compared to male-generated labour incomes, but the difference

is small. The personal characteristics of the earners do not contribute to higher shares of food expenditures, but the opposite.

In this regard, the results are more aligned with the unitary models of the household, where the identity of the income generator

does not affect substantially the household outcomes. On the other hand, household-related characteristics, such as the number

of earners and the area of residence, are the main determinants of positive changes in household food expenditures.

Apuntes de Economía y Sociedad N.º 4 (1) junio 2023/ 94-104/ ISSN: 2709 - 7005

Table 2

Regression results

95% Confidence interval

Variable

Coefficient

Robust Std. Err.

Lower bound

Upper bound

Female incomes (ln)

7.2188

1.9018

3.4897

10.9478

Female incomes (squared)

-0.5230

0.1164

-0.7513

-0.2947

Male incomes (ln)

6.8442

2.4064

2.1258

11.5626

Male incomes (squared)

-0.4984

0.1384

-0.7698

-0.2270

Food price

0.0053

0.0011

0.0031

0.0074

Average age of female members

-0.1360

0.0251

-0.1852

-0.0867

Average age of male earners

-0.0263

0.0240

-0.0733

0.0207

Years of education of female

earners

-0.9351

0.0753

-1.0828

-0.7874

Years of education of male

members

-0.8719

0.0746

-1.0181

-0.7257

Number of female earners

0.9080

0.4760

-0.0254

1.8414

Number of male earners

2.3457

0.5465

1.2742

3.4172

Housheold size (ln)

-5.0132

2.6619

-10.2327

0.2062

Area of residence (binary, rural)

2.9904

0.7360

1.5472

4.4336

Region of residence

Rest of Pacific

0.8898

0.5687

-0.2253

2.0050

Central region

-0.2565

0.5852

-1.4040

0.8910

Caribbean region

-0.3138

0.7024

-1.6912

1.0635

Number of children

-1.3151

0.3140

-1.9308

-0.6993

Household structure (binary,

nuclear)

0.4973

1.4155

-2.2780

3.2727

Constant

15.3882

13.2097

-10.5129

41.2893

Author's elaboration based on LSMS (2016a). Based on 2,981 observations.

Both female and male-generated incomes increase the share of household food expenditures, although very modestly, with

female-generated incomes having a higher effect than male-generated incomes, in line with previous research in low and

middle-income countries (Gummerson & Schneider, 2013; Hoddinott & Haddad, 1995; Levin et al., 1999). One per cent

change in female-generated incomes will change the proportion of household food expenditures by 0.0617 percentage points,

while one per cent change in male-generated incomes will alter this share by 0.0585 percentage points. Moreover, both female

and male-generated incomes will only increase the shares of food expenditures at 6.90 and 6.87 per cent, respectively,

changing to a negative relationship after those points. The results endorse the "traditional" shape of a food Engel curve -

concave downward with a positive but decreasing slope. Moreover, the effect of the food price is also very small, meaning

that the price of food explains very little of the increased demand for food led by changes in perceived incomes.

Contrarily, the personal characteristics of the household earners have a reversed effect on the share of food, but their economic

significance is minimal. Further research needs to be done to understand this relationship better, but a potential explanation

may be the highest consumption of food away from home generated by higher incomes and educational levels of the earners

(Nayga, 1996; Seidu, 2019), which is not included in the share of household food due to the lack of background information

about it. The finding is similar to those of Aguilar-Lopez, González-Andrade, and Kuhar (2020) and Brillanti, Echeverria,

and Berges (2022), but the authors do not provide an explanation for this negative relationship.

Looking at household-related characteristics, it is evident that larger households spent smaller proportions of their household

incomes in food. However, an additional male or female earner will have a positive effect over the demand for food, which is

similar to previous findings in Mexico (Djebbari, 2005), especially male household members. An additional male earner will

increase the proportion of household incomes spent on food by 2.35 percentage points, while the increase derived by an

additional female earner is only 0.91 percentage point. Next to it, there are not statistically significant differences between

nuclear dual-earner and multiearner households. This is an interesting finding that reflects the importance of the earner identity

over the earning structure where the identity of the earner is not specified.

100

Apuntes de Economía y Sociedad N.º 4 (1) junio 2023/ 94-104/ ISSN: 2709 - 7005

Finally, the share of food expenditures in rural areas is larger than in urban areas, which is in line with the existent literature

(Chikobola & Edriss, 2016; Rehman, Jian & Runqing, 2014), but there are not significant differences when comparing the

regions of residence. Residing in a rural area increases the expenditures share on food by 2.99 percentage points compared to

urban areas. For rural areas, the average of food expenditures as a percentage of household expenditures is 50.60 per cent,

and 42.58 per cent in urban areas (statistically significant difference).

V- Conclusions

This paper contributes to the literature on female-generated incomes and their contribution or effects on household outcomes

by analysing the relationship between female and male-generated labour incomes and household expenditures on food using

regression analysis. It is worth highlighting that this paper does not represent a proof for any economic model of household

behaviour.

In contrast to previous evidence for other Latin American countries, female-generated incomes do increase the share of

household food expenditures more than male-generated incomes, but the difference is minimal. A traditional Engel curve for

food is confirmed for the Nicaraguan case. The identity of the earners and the area of residence are the drivers of increased

household expenditure on food. Multiearner households alone will not generate an increase in household shares of food

expenditures, but what matters is the identity of the earners, with an additional male earner being more significant in the

increased proportion of food expenditures than an additional female earner.

While female and male-generated labour incomes are used (almost) equally for food expenditures, their use may differ for

other expenditure groups, e.g., transportation or leisure, or depending on their source, i.e., unearned incomes. Unearned

incomes carry substantial policy implications, including the effect of cash transfers and other social programmes transferring

resources into women's hands. Further research exploring whether the effect varies for unearned incomes should be developed.

Moreover, the LSMS does not allow further explore any effect of “control” over incomes on the expenditure patterns, which

can shed light to policy programas like CCT and UCT. Future lines of research could also explore other methodological

approaches that can better estimate the relationship of the variables of interest, such as estimation of food Engel curves by

quantiles. Although the methodological scope of this study is limited, its results still add policy value generating information

on earned labour incomes which could inform the linkages between the generation of incentives and policies focused on

women's insertion into the paid labour market and any expected changes in household shares of food expenditures.

Considering the strong influence of gender norms in household allocative processes and decision-making in Nicaragua

suggested in previous research, future studies should be conducted not only to account for the difference in income sources

but also for gender norms influencing household economic behaviour. Further research should also explore the negative effect

of personal characteristics on household food expenditures, since this question remains unanswered in the few studies

reviewed obtaining similar results.

VI- References

Abdullah Yusof, S., & Duasa, J. (2010). Household decision-making and expenditure patterns of married men and women in

Malaysia. Journal of Family and Economic Issue, 31 (3), 371-381. http://dx.doi.org/10.1007/s10834-010-9200-9

Aguilar-Lopez, A., González-Andrade, S., & Kuhar, A. (2020). Estimation of Engel curves for household expenditure on dry

bean and processed bean in Mexico. Acta Agriculturae Slovenica, 115(2), 215–222.

http://dx.doi.org/10.14720/aas.2020.115.2.1415

Alderman, H., Chiappori, P.-A., Haddad, L., Hoddinott, J., & Kanbur, R. (1995). Unitary versus Collective Models of the Household:

Is It Time to Shift the Burden of Proof? The World Bank Research Observer, 10(1), 1–19.

http://www.jstor.org/stable/3986564

Angelucci, M., & Garlick, R. (2014). Heterogeneity and Aggregation: Testing for Efficiency in Intra-household Allocation.

[Unpublished Manuscript].

https://pdfs.semanticscholar.org/6cf0/8a9c58566f236a83a5f70d1c7b1275a5b17d.pdf?_ga=2.55313442.806988310.

1589489800-847693442.1586881638

Armand, A., Attanasio, O., Carneiro, P., & Lechene, V. (2016). The effect of gender-targeted conditional cash transfers on

housheold expenditures : evidence from a randomized experiment. The Economic Journal, 130(631), 1875-1897.

https://doi.org/10.1093/ej/ueaa056

101

Apuntes de Economía y Sociedad N.º 4 (1) junio 2023/ 94-104/ ISSN: 2709 - 7005

Attanasio, O., Battistin, E., & Mesnard, A. (2011). Food and cash transfers: evidence from Colombia. The Economic Journal,

122(559), 92-124. https://doi.org/10.1111/j.1468-0297.2011.02473.x

Attanasio, O., & Lechene, V. (2002). Tests of income pooling in household decision. Review of Economic Dynamics, 5(4),

720-748. https://doi.org/10.1006/redy.2002.0191

Attanasio, O., & Lechene, V. (2010). Conditional cash transfers, women and the demand for food (IFS Working Paper 10/17).

Economic & Social Research Council. Retrieved from

https://ifs.org.uk/sites/default/files/output_url_files/wp1017.pdf

Attanasio, O., & Lechene, V. (2014). Efficient responses to targeted cash transfers. Journal of Political Economy, 122(1),

178-222. https://doi.org/10.1086/674968

Bhupal, G. & Sam, A.G. (2014). Female income and expenditure on children: impact of the national rural employment

guarantee scheme in India. Applied Econometrics and International Development, 14(2), 175-192.

Bircan Bodur, F. & Mukiyen Avcı, G. (2016). Household expenditure patterns: evidence from working-couple households in

Turkey. International Review of Economics and Management, 3(2), 85-108.

https://dergipark.org.tr/en/pub/iremjournal/issue/15723/293012

Bobonis, G.J. (2009). Is the allocation of resources within the household efficient? New evidence from a randomized

experiment. Journal of Political Economy, 117(3), 453-503. https://doi.org/10.1086/600076

Bradshaw, S. (2013). Women’s decision-making in rural and urban households in Nicaragua: The influence of income and

ideology. Environment and Urbanization, 25(1), 85–94. https://doi.org/10.1177/0956247813477361

Brillanti, C., Echeverria, L., & Berges, M.E. (2022, October). Consumo de alimentos y regresión por cuantiles [Food

consumption and regression by quantiles]. Paper presented at the 53 Annual Meeting of the Asociación Argentina

de Economía Agraria "Perspectivas del sector agroalimentario en la región y en el mundo", Bahía Blanca, Argentina.

Brossart, D.F., Parker, R.I., & Castillo, L.G. (2011). Robust regression for single-case data analysis: How can it

help?. Behavior Research Methods, 43, 710–719. https://doi.org/10.3758/s13428-011-0079-7

Browning, M., Bourguignon, F., Chiappori, P., & Lechene, V. (1994). Income and outcomes: A structural model of

intrahousehold allocation. Journal of Political Economy, 102(6), 1067-1096. https://doi.org/10.1086/261964

Chikobola, M. and Edriss, A. (2016) Estimation of rural-urban expenditure and elasticities of food items in Zambia: evidence

from living conditions monitoring survey. Modern Economy, 7, 567-574. http://dx.doi.org/10.4236/me.2016.75062

De Backer, K., Holvoet, N., & Milanzi, M. (2021). Intrahousehold relationships and decision-making in extended households

of the Luguru community. Journal of International Women’s Studies, 22(5), 57-71.

https://vc.bridgew.edu/jiws/vol22/iss5/4

Djebbari, H. (2005). The Impact on Nutrition of the Intrahousehold Distribution of Power (IZA DP No. 1701). Bonn: IZA.

Retrieved from http://ftp.iza.org/dp1701.pdf

Doss, C. (2001). Is risk fully pooled within the household? Evidence from Ghana.

Economic Development and Cultural Change, 50(1), 101-130. https://doi.org/10.1086/340009

Doss, C. (2006). The effects of intrahousehold property ownership on expenditure patterns in Ghana. Journal of African

Economies, 15(1), 149-180. https://doi.org/10.1093/jae/eji025

Doss, C. (2013). Intrahousehold Bargaining and Resource Allocation in Developing Countries (WPS6337). The World Bank.

Retrieved from http://documents.worldbank.org/curated/en/701071468155969077/pdf/wps6337.pdf

Elsas, S. (2013). Pooling and Sharing Incomes within Households: A Satisfaction Approach (SOEPpapers on

Multidisciplinary Panel Data Research, No. 587). Berlin: German Socioeconomic Panel Studies. Retrieved from

http://hdl.handle.net/10419/83663

Esteve, A., Castro-Martín, T., & Castro Torres, A.F. (2022). Families in Latin America: trends, singularities, and contextual

factors. Annual Review of Sociology, 48, 485-505. http://dx.doi.org/10.1146/annurev-soc-030420-015156

Esteve, A., & Florez-Paredes, E. (2018). Families in Latin America: Dimensions, diverging trends, and paradoxes. In N. Cahn,

J. Carbone, L. DeRose, & W. Wilcox (Eds.), Unequal family lives: Causes and consequences in Europe and the

Americas pp. 40-65. Washington DC: Cambridge University Press. doi: https://doi.org/10.1017/9781108235525

102

Apuntes de Economía y Sociedad N.º 4 (1) junio 2023/ 94-104/ ISSN: 2709 - 7005

Gummerson, E. & Schneider, D. (2013). Eat, drink, man, woman: Gender, income share and household expenditure in South

Africa. Social Forces, 91(3), 813-836. https://doi.org/10.1093/sf/sos173

Herrera, C., Dijkstra, G., & Ruben, R. (2019). Gender segregation and income differences in Nicaragua. Feminist Economics,

25(3), 144–170. https://doi.org/10.1080/13545701.2019.1567931

Hayashi, F. (1995). Is the Japanese extended family altruistically linked? A test based on Engel curves. Journal of Political

Economy, 103(3), 661-674. https://doi.org/10.1086/261998

Hoddinott, J., & Haddad, L. (1995). Does female income share influence household expenditures? Evidence from Côte

d'Ivoire. Oxford Bulletin of Economics and Statistics, 57(1), 77-96. https://doi.org/10.1111/j.1468-

0084.1995.tb00028.x

Instituto Nacional de Información de Desarrollo. (2016a). Encuesta de medición de nivel de vida [Data set]. Retrieved from

https://www.inide.gob.ni/Estadisticas/BDEnmv2014

Instituto Nacional de Información de Desarrollo. (2016b). Encuesta de Medición de Nivel de Vida: EMNV 2014 [Living

Standards Measurement Study: LSMS 2014]. INIDE.

https://www.inide.gob.ni/docs/Emnv/Emnv14/EMNV%202014-2%20Febrero%202016.pdf

Kazianga, H., & Wahhaj, Z. (2016). Intra-household resource allocation and familial ties. Journal of Development Economic,

127, 109-132. https://doi.org/10.1016/j.jdeveco.2017.03.002

Lesser, C. (1963). Forms of Engels functions. Econometrica, 31(4), 694-703. doi: https://doi.org/10.2307/1909167

Levin, C.E., Ruel, M.T., Morris, S.S., Maxwell, D.G., Armar-Klemesu, M., & Ahiadeke, C. (1999). Working women in an

urban setting: traders, vendors and food security in Accra. World Development, 27(11), 1977-1991.

https://doi.org/10.1016/S0305-750X(99)00096-0

Maluccio, J.A., & Flores, R. (2004). Impact Evaluation of a Conditional Cash Transfer Program: The Nicaraguan Red de

Protección Social (FCND No. 184). Washington, DC: IFPRI. Retrieved from

http://ebrary.ifpri.org/utils/getfile/collection/p15738coll2/id/76633/filename/76634.pdf

Mulamba, K.C. (2022). Relationship between households’ share of food expenditure and income across South African

districts: a multilevel regression analysis. Humanities and Social Sciences Communications, 9.

https://doi.org/10.1057/s41599-022-01454-4

Nayga, R.M. (1996). Wife's labor force participation and family expenditures for prepared food, food prepared at home, and

food away from home. Agricultural and Resource Economic Review, 25(2), 179-186.

https://doi.org/10.1017/S106828050000784X

Opata, P.I., Ezeibe, A.B., & Ume, C.O. (2020). Impact of women’s share of income on household expenditure in Nigeria.

African Journal of Agricultural and Resource Economics, 15(1), 51-64. http://dx.doi.org/10.22004/ag.econ.307616

Pickbourn, L. (2016). Remittances and household expenditures on education in Ghana's Northern Region: Why gender

matters. Feminist Economics, 22(3), 74-100. https://doi.org/10.1080/13545701.2015.1107681

Phipps, S. & Burton, P. (1998). What's mine is yours? The influence of male and female incomes on patterns of household

expenditures. Economica, 65(260), 599-613. https://doi.org/10.1111/1468-0335.00148

Qian, N. (2008). Missing women and the price of tea in China: The effect of sex-specific

earnings on sex imbalance. The Quarterly Journal of Economics, 123(3), 1251-1285.

https://doi.org/10.1162/qjec.2008.123.3.1251

Quisumbing, A. (2003). What have we learned from research on intrahousehold allocation? In A. Quisumbing (ed.), Gender

and Household Decision-making in Developing Countries: A Review of Evidence pp. 1-16. IFPRI.

Rehman, A., Jian, W., & Runqiing, Z. (2014). Estimation of urban-rural expenditure and household size elasticities of food

items in Pakistan:- evidence from PSLM survey. Asian Economic and Financial Review, 4(2), 183-190.

https://archive.aessweb.com/index.php/5002/article/view/1152

Salas Alfaro, R. y Baca Tavira, N. (2015). La contribución del ingreso femenino al bienestar del hogar en una comunidad

oaxaqueña. Espacio Abierto, 24(3), 97-113. http://www.redalyc.org/articulo.oa?id=12242627005

Schady, N., & Rosero, J. (2008). Are cash transfers made to women spent like other sources of income?. Economics Letters,

101(3), 246-248. https://doi.org/10.1016/j.econlet.2008.08.015

103

Apuntes de Economía y Sociedad N.º 4 (1) junio 2023/ 94-104/ ISSN: 2709 - 7005

Schmeer, K.K. (2005). Married women's resource position and household food expenditures in Cebu, Philippines. Journal of

Marriage and Family, 67(2), 399-409. https://doi.org/10.1111/j.0022-2445.2005.00124.x

Seidu, A. (2019). Opportunity cost of time, income, education and food away from home consumption expenditures by type

of facility in Albania. Journal of Agribusiness in Developing and Emerging Economies, 9(5), 457-475.

https://doi.org/10.1108/JADEE-06-2018-0084

Tingum, E. N., & Kuponiyi, A. (2020). The Determinants of Rural Household Food Consumption Expenditure in Lesotho:

Impact of Off-farm Income. Journal of Social and Development Sciences, 11(1), 27-35.

https://doi.org/10.22610/jsds.v11i1(S).3074

Thomas, D. (1990). Intra-household resource allocation: An inferential approach. The Journal of Human Resources, 25(4),

635-664. https://www.jstor.org/stable/145670

Thomas, D. (1993). The distribution of income and expenditure within the household. Annales d’Éconnomie et de Statistique,

29, 109-135. https://www.jstor.org/stable/20075898

Verardi, V., & Croux, C. (2009). Robust regression in Stata. The Stata Journal, 9(3), 439-453.

https://doi.org/10.1177/1536867X0900900306

World Bank. (n.d). Tasa de participación en la fuerza laboral, mujeres (% de la población femenina entre 15-64 años)

(estimación modelado OIT) – Nicaragua. Retrieved from

https://datos.bancomundial.org/indicator/SL.TLF.ACTI.FE.ZS?locations=NI

Zuniga Gonzalez, Carlos Alberto. (2010). Comparisons of LSMS-ISA data collection and dissemination efforts in Central

America. Journal of Development and Agricultural Economics, 3(8), 353-361.

https://doi.org/10.5897/JDAE.9000033

Financiación:

No se recibió ningún tipo de financiamiento.

Declaración de disponibilidad de datos:

Los datos que respaldan los hallazgos de este estudio están disponibles en Datos. Estos datos se derivaron de los siguientes

recursos disponibles de dominio público: Instituto Nacional de Información de Desarrollo (INIDE). (2016). Encuesta de

medición de nivel de vida [Data set]. INIDE: https://www.inide.gob.ni/Estadisticas/BDEnmv2014

Notes:

Encuesta de Medición de Nivel de Vida – EMNV 2014 (INIDE, 2016).

The values correspond to the category “pulperia o negocio propio” in the LSMS.

iii

It was decided to compute the natural logarithm of the household size given the difference of members among households,

where the number ranged from 2 to 21.

104