Show Summary Details

Page of

Printed from Oxford Research Encyclopedias, Economics and Finance. Under the terms of the licence agreement, an individual user may print out a single article for personal use (for details see Privacy Policy and Legal Notice).

Subscriber: Google Scholar Indexing; date: 17 April 2024

Economics and Family Structuresunlocked

Economics and Family Structuresunlocked

  • Thomas Baudin, Thomas BaudinEconomics and Quantative Methods, IESEG School of Management
  • Bram De RockBram De RockUniversité Libre de Bruxelles
  •  and Paula GobbiPaula GobbiUniversité Libre de Bruxelles


Household decisions are one of the key elements impacting many dimensions of any economy. For instance at the macro level, decisions regarding how much to save affect the economy’s investment possibilities or decisions regarding children’s education affect the overall level of human capital. Economists who study household behavior have almost solely focused on the understanding of nuclear families, i.e., parents living with their own children, childless couples, or singles. However, it is well documented that family types are heterogeneous across and within countries, both in the past and in present times. Among the different types, a classical distinction can be made between nuclear, stem, and complex households. Stem families are those allowing for three generations to live in a same household. Complex families allow for several married siblings to live together in a household. It is important to note that this is not a marginal phenomenon. For instance in China or India, the two most populous countries in the world, the majority of the population lives in either stem or complex families types. However, there is still a lot to understand about them. What are the driving factors leading individuals to form one type of family over another? Are these drivers economic or cultural? What are the intra-household dynamics of these families and how do they function? What are the implications of these differences for implementing effective family policies? The focus on nuclear families limits our capacity to answer these questions and to analyze the impact of institutional phenomena or public policies. More research to understand the determinants and functioning of other types of families hence matters both from an academic and a policy perspective.


  • Economic Development
  • Labor and Demographic Economics


Households, or more broadly families, are fascinating objects of analysis where important decisions are made. Individuals within households need to make decisions about their consumption, their participation in the labor market, their time spent on domestic work, their savings, and their investments in children. Before a household is formed, individuals also need to decide on their education, whether to marry and whom to marry, and if they married, whether they will divorce or not. Observational data show a big variety in household behavior with respect to all these decisions.

Following the pioneering works of Gary Becker, family economists have mainly used the nuclear family structure as benchmark.1 That is, a family in which parents live with their children in the household. Though important, the nuclear family is not the only way in which a family can be organized. This article claims that other forms of family structures call for more attention because the way families are organized is a crucial determinant of decisions made by households and individuals. This is particularly important because these family structures are also not static in nature, meaning that their importance and inner organization may evolve with the economic and societal circumstances.

Classifying families in function of their household organization is not a trivial task. Laslett (1972), Le Play (1884), and Todd (2011) have provided the most famous categorizations of family types, depending on their composition but also their habits after the celebration of marriage. Based on the classification of family types provided by the previous scholars, “Family Types” defines and discusses the three broadly defined family types that are considered: the nuclear family, the stem family, and the complex family. The nuclear family is a household with parents and children in which all children leave the family after marriage. This is not the case for the other two types this article considers. In complex families, children stay with the family after marriage, while for stem families only one of them remains with the parents after marriage.2

Empirical evidence shows that stem and complex families were not—and are not—marginal phenomena. They have been strongly prevalent in some parts of the world, and especially Europe, in the past (see, for instance, Todd, 2011) and they are still important in developing countries today. Using data from the Ancestral Characteristics database (Giuliano & Nunn, 2018), “Family Types” documents that, traditionally, a large fraction of the population was not living in a nuclear family in many regions across the globe. The recent Pew Research Center report confirms that this is not only a historical fact but that even in the 21st century many regions are characterized by a significant share of nonnuclear households (Pew Research Center, 2019). Why do we observe these differences in family forms is still very much an open question. Bau and Fernández (2022) review some anthropological and economic studies that tackle this question. For instance, they discuss the origins of polygyny that can be linked to agricultural technology that favors strength (Alesina et al., 2013) and the origins of complex family types that can be linked to the absence of pension plans (Ebenstein, 2014; Galasso & Profeta, 2018).3

The fact that developed countries today are mostly characterized by the nuclear family system may suggest that family structures and development are intertwined. “Stylized Facts” also provides correlations between the presence of alternative family types and economic outcomes such as GDP, inequality, and accessibility to justice. There is a strong correlation between having a higher share of nuclear families on the one hand and a higher level of GDP and accessibility to justice on the other hand. Interestingly, the opposite conclusion holds for complex families. Finally, for income inequality, we observe that the presence of a higher share of stem families is significantly correlated with a more equal society.

Hence, it is vital for economists to allow for different family types in their structural models used to understand household decisions and subsequently their impact on society.4 It is, however, clearly impossible to introduce one overarching structural model incorporating all features of the household decision process while allowing for different forms of family structure. Therefore, “Ingredients of the Structural Model” discusses the main building blocks that the literature typically takes into account when analyzing household behavior. On the one hand we have the individual preferences over all aspects in life and the household decision process to capture how these preferences are aggregated. Together this results in a household welfare function that is optimized. On the other hand, we have the different type of constraints that households face. This goes from more monetary oriented constraints, such as intertemporal budgets and household production, to more general constraints, such as marriage markets and societal context.

Subsequently, “Theoretical and Empirical Foundations for Family Types” summarizes the extensive literature, both empirical and theoretical, that has linked one (or more) of these aspects of the structural household decision models to the specific types of family organization. This complements the discussion started in Bau and Fernández (2022) and it emphasizes once more that there are no “one size fits all” answers possible. In turn, this article feeds the interesting avenue of research on improving the economic modeling of household decisions by explicitly taking up economic motivations to form nonnuclear households.

Family Types

Family structures can be categorized from the pioneering work of Le Play (1884). He proposed three different types of families; “la famille patriarcale,” “la famille instable,” and “la famille souche.” Currently, these are respectively referred to as the complex family, the nuclear family, and the stem family. Originally, Le Play (1884) defined the famille patriarcale as the most stable among the family types, where all sons remain in the family after marriage. This type of family allows customs and traditions to be directly transmitted and conserved within generations, from the elderly to the young. The opposite holds in the famille instable, where the offspring leave the parental household when they gain independence. Although Le Play (1884) acknowledged that this family type is more prone to entrepreneurship and the creation of novel ideas, he despised this structure claiming that the elderly finished their days lonely due to the excess of individualism within societies characterized by this form of family. In between these two family types, the famille souche includes one married offspring, the most capable one to continue the family business, who remains living with his parents at the family household. The rest of the offspring can stay in the household, as long as they remain single.

Other scholars have proposed more detailed and extensive classifications of family structures. Well-known ones are those of Laslett (1972) and Todd (2011). Laslett (1972, p. 31) proposes a table of five major groups of families. First, the “solitaries” include the widowed and single individuals. Second, the “no family” which are coresident siblings, coresident relatives of other kinds, and unrelated coresident persons. Third, the “simple family households” which is the nuclear family, including widows or widowers with children. Fourth and fifth are the “extended family household” and the “multiple family household” respectively. Both include some forms of stem and complex families, but the distinction is made on the number of conjugal family units: only one in the extended form, while at least two in the multiple form.

More recently, Emmanuel Todd also argues that the trilogy established by Frédéric Le Play is not sufficient to cover all the possible family arrangements around the globe, either in the past or today. Todd (2011) establishes a classification for Eurasia based upon 15 different family types. Table 1 shows this classification with the respective frequency of each type from the study of 214 populations (Todd, 2011, p. 91). He builds upon the complex, nuclear, and stem family types mainly by allowing these families to diverge with respect to their patrilocal, matrilocal, or bilocal orientations. A family has a patrilocal orientation if the bride joins the groom’s family household after the wedding, while it has a matrilocal orientation when it is the husband who enters the wife’s household. When either one or the other of these situations can occur, the family is denoted as bilocal. For instance, when a sister and a brother bring in their family house their respective spouses, the household is labeled as complex bilocal. Todd’s analysis however suggests that the patrilocal orientation dominates in complex and stem family types.5

Table 1. The Fifteen Family Types According to Todd (2011)

Family type

Frequency (%)

Complex patrilocal


Complex matrilocal


Complex bilocal


Stem patrilocal


Stem matrilocal


Stem bilocal


Nuclear, integrated patrilocal


Nuclear, integrated matrilocal


Nuclear, integrated bilocal


Nuclear with temporary coresidence patrilocal


Nuclear with temporary coresidence matrilocal


Nuclear with temporary coresidence bilocal


Nuclear egalitarian


Nuclear absolute


Stem with additional temporary coresidence


The ultimate goal of this article is to convince economists to properly acknowledge the existence of nonnuclear family types and to develop new structural models to grasp the impact of the different family types. In this respect the focus of the article will only be on three broad types: the nuclear family, the stem family, and the complex family. Following Laslett (1972, p. 41), Figure 1 provides some examples of ideographs to further illustrate these three types. These ideographs are commonly used in anthropology to represent families. The household illustrated in Figure 1 shows on top a couple with three children, two sons and one daughter. The ideograph in the middle represents a stem (patrilocal) family with three generations living in the same household. The younger couple has a son and a daughter. The last ideograph depicts a complex (patrilocal) family where the two sons of the first generation live in the household with their respective wives and children.

Figure 1. Ideographs for family structures.

Stylized Facts

This section first documents the heterogeneity of family types around the world in both the past and the present. Subsequently, it presents some correlation results between ancestral family organization and current economic outcomes related to GDP, inequality, and justice accessibility. All this motivates the clear need to extend the economic models to allow for nonnuclear families.

Heterogeneity of Family Types Around the World

Figures 24 show the current prevalence of ancestral nuclear, stem, and complex family types around the globe. The figures use data from the most complete version of the Ancestral Characteristics database (Giuliano & Nunn, 2018). These data provide worldwide country-level measures for the share of the current population that has a given cultural or environmental preindustrial ancestral characteristic. The variables of the Ancestral Characteristics database are those of the Ethnographic Atlas, while the database is enriched with ethnic groups of Eastern Europe, Siberia, and European groups that are not present in the Ethnographic Atlas.

The Ethnographic Atlas is an ethnicity-level database for 1,265 different ethnic groups around the world constructed by George Peter Murdock (Murdock, 1967). Several scholars have used the digitized version of the database, released in 1999. Lowes (2021) provides a survey of papers that have used the Ethnographic Atlas, and a discussion on the limitations of these data. Information on the ethnic groups comes mostly from the 19th and early 20th centuries and, in particular, prior to European contact for African and American countries.

Figure 2 shows the fraction of a country’s population (with nonmissing ancestral data) with ancestors organized as nuclear, monogamous families. In detail, this corresponds to the variable


of the Ancestral Characteristics database. The variable v8grp1 reports the fraction of a country’s population for which data on the prevailing form of domestic or familial organization is missing. The variable v8grp2 is defined as the fraction of a country’s population with ancestors organized as “independent nuclear families with monogamy.” Next, Figure 3 shows the fraction of a country’s population with ancestors organized as a stem family type. This corresponds to the variable


in the Ancestral Characteristics database. This variable is defined as the fraction of a country’s population with ancestors organized in “minimal extended or ‘stem’ families, i.e., those consisting of only two related families of procreation (disregarding polygamous unions), particularly of adjacent generations.” Finally, Figure 4 shows the fraction of a country’s population with ancestors organized as complex families. This corresponds to the variable


in the Ancestral Characteristics database. v8grp8 and v8grp9 are respectively defined as the fraction of a country’s population with ancestors organized in:

Small extended families, i.e., those normally embracing the families of procreation of only one individual in the senior generation but of at least two in the next generation. Such families usually dissolve on the death of the head,” and “Large extended families, i.e., corporate aggregations of smaller family units occupying a single dwelling or a number of adjacent dwellings and normally embracing the families of procreation of at least two siblings or cousins in each of at least two adjacent generations. (see the description of variable EA007 in the Ethnographic Atlas)

The pattern displayed by these figures is clear. The nuclear family ancestral trait clearly prevails in developed countries. The complex family trait is salient in Iceland, New Zealand, and Asian countries, in particular in Kyrgyzstan, Tajikistan, Uzbekistan, and Turkmenistan. Most African countries display the complex family trait. This trait can also be observed in Latin American countries, for which we encounter both nuclear and complex ancestral family types. Finally, the stem family trait is relevant in Japan, Ukraine, Belarus, Poland, Slovakia, the Czech Republic, and Portugal.

Within and beyond national frontiers, family types can also be illustrated at the ethnicity level. For this, we can combine the Ethnographic Atlas with the mapping of historical ethnicity boundaries provided by Murdock (1959). Figure 5 shows the diversity in family types across ethnic groups in Africa. The figure shows indeed that ethnic frontiers allow to capture a wider level of heterogeneity not only between countries but also within countries.6 Figure 5 also shows “the prevailing form of domestic or familial organization” as defined in the Ethnographic Atlas.7

Data from the Ancestral Characteristics database show that current populations over the world do not all share the same ancestral family characteristics but they remain silent about the current prevalence of each type of family. In a recent report, the Pew Research Center (2019) shows that the organization of families remains quite heterogeneous, even today. In China, India, and Senegal, the majority of the population lives in extended-family households. This is also the case for Liberia, Tajikistan, Nepal, and Namibia. An extended-family household is defined as “a household that includes relatives other than children or partners. For example, adults who live with their siblings or parents in addition to their children.” Defined like this, extended-family households include stem and complex families. Although the proportion of these family types seems negatively associated with the level of development, it remains noticeably high in rich countries like Germany and Norway with a share of 17%. At the world level, 38% of individuals and 26% of households live in such extended-family households. Including individuals in polygamous households increases the estimation to 40% (and to 26.5% for households). Hence, the nuclear family is far from being the only important type of household today, urging economists to better model family decisions in nonnuclear settings.

Figure 2. Fraction of the population in nuclear (monogamous) families.

Source: Ancestral Characteristics database Giuliano and Nunn (2018).

Figure 3. Fraction of the population in stem families.

Source: Ancestral Characteristics database Giuliano and Nunn (2018).

Figure 4. Fraction of the population in complex families.

Source: Ancestral Characteristics database Giuliano and Nunn (2018).

Figure 5. Family types in Africa: Nuclear, polygynous, stem and complex families.

Family Types and Economic Outcomes

Tables 24 illustrate existing correlations between these ancestral traits of family types and current economic outcomes. Formally, we estimate with ordinary least squares (OLS) the following equation model:

yi=βFi+xiγ+Ei (1)

where i denotes a country, and y is an economic outcome. The economic outcome can either be the GDP per capita in 2010, the average GINI between the years 2010 and 2020, or the rule of law index for the year 2020. GDP per capita and the GINI index are part of the World Development Indicators data made available by the World Bank. The rule of law index is created by the World Justice Project and provides a measurement for the accessibility to justice of citizens, whether people have access to courts or whether crime is effectively controlled. Fi is the family type: the fraction of a country’s population with ancestors organized in either nuclear, stem, or complex families. These three variables were defined in detail in “Family Types” and illustrated in Figures 24. Finally, x are geographic controls for the ancestral distance from the coast and terrain ruggedness, also provided in the Ancestral Characteristics database. As explained in Giuliano and Nunn (2018) and in line with previous studies which focused on the historical importance of geography on current outcomes, ancestral geographical characteristics show a strong association to current economic outcomes (see, among others, Michalopoulos, 2012; Fenske, 2014; Henderson et al., 2018).

Results shown in Tables 2 and 4 suggest that countries in which the nuclear family ancestral characteristic was more prevalent do better in terms of GDP per capita and rule of law. The opposite holds for countries in which complex families prevailed. Table 3 suggests that inequality is negatively associated with the presence of stem families in the past. Countries in which stem family types prevailed are associated with a lower GINI index, hence more equality. Taking the nuclear family as example, the exact magnitudes of the coefficients can be interpreted as follows. Controlling for geographic characteristics, countries where nuclear families fully prevailed (such as Denmark, the United Kingdom, or Uruguay) have on average a higher GDP per capita in 2010 (i.e., 21,099 USD), a higher GINI coefficient (i.e., 0.0766), and a higher rule of law index (i.e., 0.110) compared to countries with no nuclear families as ancestral family trait (such as Afghanistan, Pakistan, or Jordan).

Table 2. GDP and Family Types

Dependent variable is GDP in 2010













Geographical controls




Dependent variable:

Mean [standard deviation]

15,352.18 [22,025.08]









Note: Standard errors in parentheses. ***p < 0.01. Geographical controls include distance from the coast and terrain ruggedness.

Table 3. Inequality and Family Types

Dependent variable is GINI (2010–2020)













Geographical controls




Dependent variable:

Mean [standard deviation]

38.14 [7.70]









Note: Standard errors in parentheses. ***p < 0.01. Geographical controls include distance from the coast and terrain ruggedness.

Table 4. Rule of Law and Family Types

Dependent variable is rule of law in 2020











-0.0711** (0.0308)

Geographical controls




Dependent variable:

Mean [standard deviation]

0.549 [0.141]









Note: Standard errors in parentheses. ***p < 0.01,

**p < 0.05,

*p < 0.1. Geographical controls include distance from the coast and terrain ruggedness.

In general, these results are in line with the following quote from Cox and Fafchamps (2007):

Extended families are important just about everywhere, but especially so in poor countries, where social safety nets are incomplete or nonexistent and households must cope with an unforgiving environment of severe poverty and shocks to economic and physical well-being. Autonomy is not a likely option for a household struggling to make ends meet in the face of looming disasters such as drought, flooding, pestilence or infectious disease—especially against a backdrop of inadequate formal credit and insurance markets and a minimal welfare state. In poor, laissez-faire economies ties to communities, friends and relatives—both near and far—can make the difference between surviving and perishing.

Importantly, this quote also illustrates that the associations showed in Tables 24 can capture impacts running in both directions.

More in detail, results for GDP and the GINI index are in line with those of Duranton et al. (2009), who provide a similar analysis for the nomenclature of territorial units for statistics (NUTS) III European regions with data on family types from Todd (1990).8 Importantly, they find that family structures are very strongly related to economic outcomes, such as education attainment, income per capita, growth, inequality, or employment. More specifically, they find that regions with absolute nuclear families tend to have more educated individuals, higher employment rates, and higher GDP per capita.9 Communitarian/extended family types tend to be associated with poorer societies, mainly manufacturing, and with lower inequality.

These results hence join the view of Duranton et al. (2009) that past family types can have a strong impact on the present-day level of development of societies. In addition, the rule of law index provides an interesting outcome when it comes to how families affect societies, as families can often replace poor institutions. This is also aligned with Guirkinger and Platteau (2020), who argue that “the family is essentially viewed as a substitute for markets, legal enforcement mechanisms, and state-devised social protection.”

Ingredients of the Structural Model

This section provides a list of the main ingredients for any type of household decision-making model. “Theoretical and Empirical Foundations for Family Types” provides some reflections on how either nuclear, stem, or complex family types can be more or less related to these ingredients. The choices made by the household are governed by the household welfare function, which captures two different things: the individual preferences of the household members and the household interaction process to aggregate these preferences. Next there are a series of constraints that the household is facing: the intertemporal budget constraint, the household production process, the marriage market, and the societal context (including institutions and norms).

The Individual Preferences

It is important to allow for heterogeneity in preferences across different household members (see, e.g., Croson & Gneezy, 2009 for a recent discussion on gender differences). This is captured by the individual utility function. A first argument of this function is consumption goods. Those can be home produced or market goods. As introduced in Browning and Chiappori (1998), these goods can be partly private (e.g., clothing) or partly public (e.g., housing) in nature. A second argument is leisure in order to capture the important dimension of time-use decisions (see, e.g., Apps & Rees, 1988 for a discussion on proper welfare analysis in a household context). Time-use decisions capture both labor market decisions (including agricultural production) and household production decisions. These two arguments together are basically the main ingredients of any labor supply model.

A third argument, stemming from a different literature, is related to social norms and can be introduced in the spirit of Lindbeck et al. (1999). That is, agents may suffer a disutility when not following the prevalent social norm. For instance, if the norm is to form a stem family, forming a nuclear family would incur a social punishment in the form of a utility loss. This loss is an increasing function of the share of the population respecting the norm. In the presence of social norms, families may have to decide whether to follow them or not, implying the comparison of expected (indirect) utilities.

Next, preferences also reflect intergenerational altruism and cultural transmission. There are many ways to model dynastic altruism in the context of family economics (Becker & Barro, 1988; Gobbi & Goñi, 2021; Razin & Ben-Zion, 1975; etc.). When exploring cultural transmission, the privileged representation is paternalistic altruism by Bisin and Verdier (2001). In this framework, altruism is imperfect as decisions made by offspring are evaluated through the lens of parental preferences. Because of this, parents expect their children to be better off if they adopt their own preferences. A similar reasoning also implies that children are often considered as a “public good” in which parents invest time and money (Blundell et al., 2005; Cherchye et al., 2012; Gobbi, 2018).

Finally, several of the arguments of the utility functions may be sensitive to (unexpected) shocks such as job loss, death, bad harvest, etc. To capture this, preferences may be deterministic or stochastic in order to take risk attitudes and uncertainty into account. Naturally, they can also be static or intertemporal to model forward-looking behavior (see Chiappori & Mazzocco, 2017 for a recent review).

The Household Welfare Function

The individual preferences of the household members can be selfish or take other aspects such as altruism or externalities into account. Related to this, the arguments can enter based on the individual share of consumption (e.g., the mother’s share of clothing expenditures) or as the total amount of consumption (e.g., housing or total expenditures on clothing). All this implies that there may be many different ways how the household members interact to reach a joint household decision; see Browning et al. (2014) for a recent review and Baland and Ziparo (2018) for a critical discussion in the context of nonnuclear families.

In line with the seminal work of Gary Becker (1981), the decision process can be dictatorial, meaning that the household can be represented as a single decision-maker. These decisions can then reflect only the preferences of the selfish dictator (who is basically only taking his own preferences into account) or those of the paternalistic dictator who is aggregating the preferences of the individual members. Importantly, however, this decision process implies that the aggregation remains constant over time and is for instance not depending on shocks or changing environments. The latter is not what is typically observed in the data.

Richer structural models allow for a household decision process that may take strategic consideration or shocks into account. Strategic considerations may be the result of free-riding behavior (e.g., investments in public goods), intertemporal concerns (e.g., establishing reputation or absence of commitment), or principal agent problems (e.g., actions cannot be fully monitored). Collective models, after Chiappori (1988), or more generally cooperative models, assume that the members can circumvent these concerns and can reach a Pareto optimal decision (i.e., no household member can be made better off without making another worse off). A motivating argument for this approach is the repeated interaction in households or the importance of quality of marriage. Noncooperative models meanwhile will focus on a Nash equilibrium, meaning that household members base their decisions on their beliefs of the actions of the other members. This will typically result in a decision that is not Pareto optimal (e.g., due to the underprovision of public goods), which implies that all household members could be better off with an alternative decision.

In all these models, even if their preferences enter the decision process through dynastic altruism, children are not considered as decision-makers. Also, the models will typically only allow explicitly for two decision-makers (i.e., the parents). In the context of nonnuclear households, or more generally for households with older children, this may not be the most realistic choice to properly capture the heterogeneity in the household decision process.

The decision process in combination with the individual preferences form together the household objective function. This is the function that is maximized by the household, meaning that it governs the household decisions. Economists will in turn use this construct in their counterfactual analysis to predict behavioral reactions in new situations.

Intertemporal Budget Constraint and Household Production

The most natural constraint on the household choices is the budget constraint, which includes expenditures in both time and money. Income can be earned through labor market decisions, the result of borrowing/saving decisions, or be available through endowments (e.g., inheritance, available time). To model bequest motives, bigger decisions (e.g., housing), or unexpected income shocks, it is often important to have an intertemporal budget setup.

Another source of income is household production. The household, or more broadly the family, can be viewed as a small firm that produces several goods. In developing contexts, it is important that this includes agricultural production, both for own consumption and for selling, since this is a main source of income. Besides agricultural production, other goods that are typically considered are household chores and children.

Clearly individuals will have preferences over the outcome of this production process but at the same time it can be considered as a constraint. To be more specific, there are clearly economies of scale related to the production process (e.g., all members benefit from the chores, risk sharing in agricultural production) but there could also be room for economies of scope or even specialization (e.g., experience in child-related activities). Moreover, in the case of agricultural production, it is important to acknowledge that assets may be essential to have access to this source of income. All this implies that the production process may be restrictive in, for instance, the partner’s choice or the place to live.

Marriage Market and Societal Context

A different set of constraints that the household is facing comes from the marriage market (e.g., free partner choice, dowry, bride-price) and the societal context (e.g., poor institutions, norms, religion). These are in principle not necessarily monetary related but can clearly severely restrict the choice set, impact the production possibilities, or decrease welfare due to violations of norms.

Moreover, as discussed at length in Browning et al. (2014), all this can also have a significant impact on the household decision process. For instance, the starting point of the richer structural models is often that individuals have outside options through the support of the family or divorce. The validity of this approach thus crucially depends on how the marriage market functions or what the social context allows for. This has been well investigated by the literature and the section Theoretical and Empirical Foundations for Family Types discusses many concrete examples.

Theoretical and Empirical Foundations for Family Types

This section links the ingredients of the structural model to the key takeaways from the vast (mainly empirical) literature on family types. The aim of the discussion is to show that some components of family decisions might be more salient in certain types of families than in others. Be it in terms of the decision structure of the family or the constraints they face. Table 5 summarizes these associations. This section therefore aims at guiding, both the empirical and theoretical, researchers in developing the right framework for the specific setting at hand.

The Individual Preferences

We begin the reflection on particular types of preferences and their associations to family structures. First, those associated with social norms, and second, those reflecting altruism. A “taste for privacy” could also be added to the list, such as in Foster and Rosenzweig (2002) or Kaplan (2012). That would push households to be nuclear, which does not require a discussion. This taste for privacy could be thought of in terms of a luxury good, that would give rise to more nuclear families in richer societies (as in Table 2).

Social Norms and Cultural Transmission

When thinking about the intrication between social or cultural norms and family types, it may be worth distinguishing two questions. First, how do current norms influence family forms and the behaviors within them? Second, how are these norms transmitted from one generation to another? Following the representation of social norms by Lindbeck et al. (1999), the nuclear family is probably the less efficient form for what concerns the enforcement of social and cultural norms as they have less old adults monitoring children and young adults’ behaviors. As a result, nuclear families are also the more adaptable to changes in social norms. Hence, inefficient social norms are more easily abandoned than in any other family structures. Within complex families, the enforcement of norms is better accomplished as each family member can monitor and report to the household head. A middle ground is found in stem families, with some monitoring, but less than in complex families, and a better ability to abandon inefficient social norms.

With regard to how these norms are transmitted between generations, the seminal contribution by Bisin and Verdier (2001) can be used.10 Family is the first place of socialization and so the place where norms and culture pass down from parents to children. Society outside the family is only the second place of socialization. Parents invest in costly socialization effort to pass their preferences, culture, or internalized norms to their children because they are “paternalistically altruistic” (see “Altruism and Family Ties” for more discussion). Surprisingly, Bisin and Verdier as well as the rest of the literature only focus on nuclear families. So, there is no study on how cultural transmission, especially the transmission of norms, is influenced by family structure, while this is one of the core messages of Le Play (1884): stem and complex families are better suited to perpetuate traditions and respect of the elderly.11 Further research is called for in this area but some intuitions can be given. There are mainly two forces opposing family types as regards socialization and the perpetuation of traditions. In nonnuclear families, there are more persons to monitor the socialization process of the younger generation and to report any misconduct to the head of the household than in nuclear families. It implies that, if the main vectors of socialization, the elders, agree on the norms and culture to be passed (they are culturally homogeneous), they have a cultural advantage on nuclear families due to their number. This is the first main force. But conversely, it is potentially harder to reach a consensus on the norms and cultural traits to pass to children when families are complex; the risk of disagreement is larger and so, the signals sent may be contradictory. In that case, nuclear families have a cultural advantage in the socialization process compared to other families as the probability to reach an agreement is higher.

Altruism and Family Ties

Bisin and Verdier (2001) explain why inefficient norms or cultures may persist over time thanks to their concept of paternalistic altruism. With this kind of imperfect altruism, parents value the future choices of their children through the lens of their own preferences. It follows that parents try to transmit their own preferences and norms to their children and so contribute to perpetuate potentially inefficient societal and cultural contexts. This result is evidenced, for instance, by Chabé-Ferret (2019) in the context of fertility decisions of second-generation migrant women in France and the United States. Women originating from high-fertility countries tend to shorten birth spacing compared to natives even in situations where such shortening is suboptimal.

That being said, altruism does not need to be imperfect to influence the distribution of family types and behaviors within families. A high degree of (perfect) altruism toward siblings and children should be negatively associated with the persistence of inefficient social and cultural norms. Indeed, a higher degree of altruism should incite people to adopt family forms which are efficient for their relatives. It implies that in a context where public institutions protect family members inefficiently, complex forms of family should be more prevalent; while good institutions should reduce the incentive to form complex families. A high degree of altruism should magnify this effect. To the best of the authors’ knowledge, there is no study of the association between the degree of altruism and the prevalence of family types. Nevertheless, this question can be looked at through the lens of family ties as the intensity of family ties is positively associated with intrafamily altruism (Alger & Weibull, 2008).

In their key contribution, Alesina and Giuliano (2015) propose a review of the literature on family ties and their importance for the economy. They show that when family ties are strong, the family constitutes the sphere of trustable persons, which translates into decreased civic sense and lower levels of trust outside the family. It goes hand in hand with high levels of home production performed, most of the time, by dependent members of the family. On the contrary, when family ties are weak, people have a better sense of civil responsibilities and a higher sense of equality within households. As a matter of result, nuclear families, which are characterized by more independence between children and parents, are associated with weaker family ties than stem and complex families which give more power to the family.

In a famous and debated article, Reher (1998) divides European families on the basis of the intensity of their family ties. He shows that European families can be divided into two main categories: the weak families in the North and the strong families in the South. In between, countries like France and Germany are more difficult to characterize. Strong families, like the Italian and Spanish ones, are characterized by a higher level of control inside the family sphere but also a higher degree of cohesion and solidarity, compared to weak families in the North. Reher (1998) provides a series of striking differences between these two main regions. For instance, homelessness prevalence is strongly and positively associated with the prevalence of weak families. Reher’s (1998) definition and analysis of family ties is not radically different from the one proposed by Alesina and Giuliano (2015) but his conclusion is that the mapping between Todd’s types of families and the distinction between strong and weak families is far from obvious: where Todd’s family types are diverse over small territories, Reher’s family types are more homogeneous. It remains however that, at large, strong families are more often complex than nuclear while weak families are more often nuclear.

Finally, it is important to note that altruism also determines the way discounting is introduced in the individual preferences. Galperti and Strulovici (2017) distinguish between direct pure altruism, in which the utility of the first generation directly depends on that of all future generations, and indirect pure altruism, in which the utility of the first generation only indirectly depends on the utility of future generations through the utility of the second generation. These authors show that direct pure altruism implies time inconsistency, while indirect pure altruism implies time consistency. For instance, the seminal paper by Becker and Barro (1988) assumes indirect pure altruism, while Gobbi and Goñi (2021) assume direct pure altruism in the context of the British aristocracy, which is characterized by stem family types. Direct pure altruism could be more associated with complex or stem families, in which the future of the family has a (direct and) strong weight, than the one of nuclear families.

The Household Welfare Function

The second major ingredient of the household welfare function is the decision process within families. Which type of family type could be better modeled with a unitary decision model rather than a collective one? Or with a noncooperative one? These questions have been largely discussed in the context of nuclear families (for reviews, see Browning et al., 2014 and Doepke & Tertilt, 2016). This section summarizes the evidence suggesting that some family types tend to favor one type of decision process versus another.

First of all, altruism as well as family ties not only alter the distribution of family types but also potentially shape the behavior inside each kind of family. For instance, using modern evidence from the Bamileke ethnic group in the city of Bafoussam in the West region of Cameroon, Baland et al. (2017) study the pattern of transfers within complex families in the context of limited access to credit or saving facilities. They focus on transfers between siblings, which represent the majority of all transfers within the family. They show that when individuals are young, the older siblings transfer to their younger siblings. The direction is reversed when these grow older and it is the younger siblings’ turn to transfer to their older siblings (and their offspring). Another example is given in Edlund and Rahman (2005) who compare the influence of a father—which is assumed to be stronger in nuclear families—versus that of a grandfather—assumed stronger in complex or stem families—on children’s education and health outcomes. They find that children’s education tends to be higher in nuclear families than in complex families and find no differences in health outcomes, proxied by the height-to-age. Related to the previous paper, Le Bris (2020) provides evidence that stem and complex families are associated with strong parental authority while children enjoy more freedom in nuclear families. He finds that strong parental authority has a positive effect on human capital investment. A second dimension is related to public goods and, more generally, monitoring actions of household members. On the one hand, the bigger the family is, the higher the economic gains of sharing public goods are and the more the cost of the public good can be spread out through family members. On the other hand, free-riding and moral hazard concerns arise more easily in larger families. The complex family type has of course the highest number of players and negotiators and so potentially the highest incentive to adopt a unitary (dictatorship) decision framework. Any noncooperative negotiation process will imply a lot of inefficiency. Therefore, we might see more often a strong decision power of one of the family members. Moreover, the cost of monitoring is probably lower (since it can be done by more people), meaning that we observe less strategic behavior. Societies with complex families are probably also more traditional meaning that the unitary model might be more salient (i.e., dictatorship). Recently, Rangel and Thomas (2019) have indeed argued that, in the context of complex households in Burkina Faso, the allocation of resources was consistent with Pareto efficiency. However, this view can be debated; Baland and Ziparo (2018) suggest that nuclear families tend to show stronger cooperative behavior. In a famous paper, Anderson and Baland (2002) also showed that noncooperative behavior between husband and wife could explain saving decisions in a context where complex families were common.

Finally, domestic violence also correlates with the household decision process to some extent. Using data from historical Spain, Tur-Prats (2019) shows that, compared to nuclear families, stem family types have a causal negative effect on domestic violence. She establishes causality using an instrumental variables strategy that exploits exogenous variation in inheritance laws that followed the Christian conquest of the Iberian Peninsula. Among the possible mechanisms, she suggests that: (a) the presence of more witnesses can refrain a violent husband from acting, and (b) since domestic tasks are also ensured by the mother-in-law, women can increase their labor force participation, and hence their bargaining power. Lowes (2018) also provides another important example in which the family structure can affect the decision process. She shows that matrilineal kinship systems reduce cooperation between spouses (compared to patrilineal ones). This is however beneficial to women themselves, who face less domestic violence, and their children, who benefit from larger investments.

Intertemporal Budget Constraint, Household Production, and Family Types

We now discuss how production activities and budget-related considerations can affect family forms and vice versa.


We begin by reviewing the contributions linking family types to different types of production activities. Pensieroso and Sommacal (2019) provide a theoretical model to explain how the structural transformation from agriculture to industry was a determinant for the aggregate shift from stem to nuclear families in the United States between the 19th and 20th century. Importantly, their model builds upon Pensieroso and Sommacal (2014) where intergenerational coresidence is determined by the income of the young adults relative to that of the elderly. To the authors’ knowledge, their model is the first to take the family transformation process as endogenous in a (macro) structural model.

An important production factor associated with family types is the land to labor ratio. Voigtländer and Voth (2013) claim that the high land to labor ratio, caused by the Black Death, was key for the emergence of the so-called “European marriage pattern” (EMP). As described by Hajnal (1965), the EMP combined late marriages (above 23 for women and above 26 for men) and high lifelong celibacy rates (20%).12 These demographic patterns were essential to ensure both population and economic stability in preindustrial times.13 Voigtländer and Voth (2013) show that the Black Death, by favoring pastoral agriculture and then husbandry, has reinforced the role of women in the economy and provoked at least partially the emergence of the EMP. The argumentation of Voigtländer and Voth (2013) for the emergence of late marriages could also apply to the emergence of nuclear families, as mentioned in Guirkinger and Platteau (2020).

Family types, or more generally societal norms, can also affect the way production activities are accomplished. For instance, Fafchamps (2001) provides a theoretical framework to analyze household production and intrahousehold equity in complex families. Krishnan and Sciubba (2009) study how extended families facilitate the formation of labor pooling groups. Finally, the seminal paper by Udry (1996) shows that existing norms, preventing husbands to work on the land of the wives, leads to inefficient household decisions.


The family is also an institution capable of providing insurance to its members:

The family—or more accurately, the kinship group—is important in traditional societies in large measure because it protects members against uncertainty.

(Becker, 1981, p. 343)

Following this logic, the larger, complex family can be an organizational form that provides individuals a better form of living in societies where one has to cope with high levels of uncertainty. Uncertainty can take several forms, such as uncertainty associated with income losses, unemployment, or health shocks.

The role of the family in the context of uncertainty associated with income losses has been studied by Foster and Rosenzweig (2002). They provide a structural model of complex family breakups, building upon the collective choice model. In their model, families gain from coresidence from sharing the cost of a household-specific public good and from sharing information regarding farming techniques. Coresidence will be desirable given a certain economic environment, which includes scale economies in production and risk sharing. The parameters of the model are estimated using Indian data over the period 1971–1982. The authors find that households are better off coresiding than splitting into nuclear families in environments where income risk is high. Pressure for risk sharing among household members can however lead individuals to hide their income from one another in order to avoid being asked for money by other members of the household. Kinnan (2021) shows indeed that hidden income is the main reason for incomplete insurance in rural Thailand.

Next, health shocks are another important risk element that any individual faces. The risk of having to undertake medical expenses is an important reason why singles and couples save after retirement (De Nardi et al., 2021). Being in a couple allows one to rely on spousal caregiving, and hence (all else equal) diminish the amount of savings per individual. By extension, complex families can reinforce this insurance device. Related to this, in the context of prerevolutionary France, in the Pyrénées where stem family types prevailed, some authors mention the existence of the droit à la chaise (right of a chair). This implicit right goes to the nonheirs of stem family types, who would be allowed to return to the family house when they are elderly and have a secured place next to the fireplace (Zink, 1993, p. 173). Finally, families also play an important insurance mechanism against unemployment. Kaplan (2012) shows that the possibility for young adults to move back to their parents’ house represents an important form of insurance against unemployment in the United States. In turn, the possibility of coresidence allows young adults to search longer for better jobs.14

Intertemporal Components: Inheritance

Finally, turning to the intertemporal components of the family budget, it is worth mentioning the role played by inheritance customs and laws in shaping family forms. An important reason for the existence of families relates to the transmission of goods and assets throughout the same lineage. As has been extensively discussed in other social sciences, the family structure is highly dependent on the inheritance scheme that is used in a society (Berkner & Mendels, 1978; Goody et al., 1978; Le Roy Ladurie, 1972). The reason why different inheritance schemes relate to different family structures is simple. As noted by Habakkuk (1955), any form of inheritance shapes the nature of relationships within a family—between parents and children, between siblings, or between husbands and wives. Given these different relationships, diverse family structures arise—different types of coresidence (inter- and intragenerational), different marriage arrangements, different fertility decisions, or different investments in children of different gender. For example, on the one hand, partible systems of inheritance, which divide family wealth among offspring upon inheritance, are linked to nuclear families. On the other hand, impartible inheritance, which prevents the family wealth from splitting by assigning it to a single heir, is commonly linked to stem family types. The reason behind these differences when it comes to the preindustrialization period in Europe was that receiving an inheritance allowed individuals to form a household and hence enabled them to marry. Exclusion from inheritance largely increased the probability of celibacy (Bonfield, 2001; Bourdieu, 2002; Zink, 1993). In economics, Bertocchi and Bozzano (2015) find that family structure, which they proxy by both “residential habits” (nuclear and complex families) and inheritance schemes (partible inheritance and primogeniture), is a major driver of the education gender gap over the period 1861–1901 in Italy. They find that nuclear families and partible inheritance are associated with a higher female-to-male enrollment rate ratio in upper primary schools. The authors suggest that the reason for this finding relies on the fact that nuclear families are more liberal and less authoritarian and partible inheritance implies more equality among offspring compared to an impartible system, such as primogeniture. Also linking inheritance to family types, Le Bris (2020) shows that systems of inheritance that favor inequality among children lead to greater investment in physical capital. The reason is twofold. First, inequality maintains wealth large enough to allow for investments to happen (while systems of inheritance that divide family wealth lead to portions that are too little to invest). Second, investments are more efficient because they are not constrained by an equal sharing inheritance rule and can therefore consider nondivisible assets.

Wealth and income are important aspects of whether individuals can move from stem or complex families to nuclear ones. In some places, owning a house is restricted to a few heirs while the remaining of the family has no possibility of buying a home. This was the case for instance in the region of the Pyrénées around the year 1760 (Zink, 1993, p. 282). There, no possibility of buying wood for construction existed and hence there was no possibility of building new houses. This maintained the family united within the same house and the stem family type prevailed (nonheir siblings who remained in the family house were very unlikely to marry).

Also building upon the strong connections between family structures and inheritance, Galasso and Profeta (2018) propose a two-period overlapping generation model with voting that explains how family culture, proxied by different inheritance rules, can be an important determinant for the adoption and generosity of public pension systems. Their findings suggest that pensions emerge to replace private family transfers. The level of generosity is however different between impartible inheritance societies and partible inheritance ones. In impartible inheritance systems, a basic, minimal, pension system emerges while a generous pension system emerges in partible inheritance societies.

Housing and Land Scarcity

From a development perspective, increasing levels of land scarcity has several implications. Guirkinger and Platteau (2015) provide a theoretical model that accounts for the fact that growing land scarcity increases the individualization of family farms, hence leading to more nuclear families and fewer complex ones. In their model, this process is the result of a two-stage maximization problem. In the first stage, the head chooses how much to consume, how much to give to the members of the complex family, and the size of it. In the second stage, family members individually decide upon how much effort to put into farming, both for the complex and the nuclear family. Alston and Schapiro (1984) have also argued that the role of inheritance, and hence family structures, in shaping family decisions might be dominated by the role played by land availability.

Looking at contemporary data for Europe, Peters et al. (2021) show that countries where home ownership is high also display higher levels of intergenerational coresidence. They propose a model where cohabitation provides an informal credit and rental market at the household level. The benefits from cohabitation are higher when young adults have low income or when the formal rental and credit markets have higher frictions.

Institutions, Marriage, Religion, and Family Types

Stylized Facts” showed that the rule of law was positively related to societies with high nuclear family ancestral traits and negatively related to high complex family traits (Table 4). This section begins with a review of studies providing a theoretical basis for such relationships.

In the past, social security was provided by a large extended form of family (a clan, a lineage, a tribe). Greif (2006) explains how the emergence of Western corporations in medieval Europe were complementary to the existence of nuclear families and the decline of large kinship groups. These corporations provided social safety nets against famine, unemployment, and disability, previously provided by the kinship group. In turn, the nuclear family fostered economic growth through later marriages and lower fertility rates.

The effect of past family systems goes well beyond pension systems. Alesina et al. (2015) show that medieval family structures in Europe are strongly associated with the contemporaneous desire for regulation. In a nutshell, people originating from regions where extended and communitarian families were dominant tend to adhere to more pro-job security than people originating from regions where the nuclear family was dominant. This means that current rigidities on labor markets originate potentially from the family structures prevailing in the past. Relying on the epidemiological approach, Carta et al. (2021) link the voting behaviors of American residents to their cultural background regarding family structures. They show that descendants of regions where families were egalitarian vote for more generous childcare programs, which in the end has a significant effect on childcare policies implemented by U.S. federal politicians.

Next, the article discusses how religion and marriage might have affected family types. Because religions decree codes of conducts within and outside families, they have long been crucial determinants of family forms, especially in Europe. Building on Goody and Goody (1983), Schulz et al. (2019) explain how the Christian Church implemented specific policies to ban marriage practices that were commonly used to enforce alliances between families.15 The main goal of the Church was to limit the power of kinship groups. Since medieval times, incest has been more and more punished and its definition extended to always more distant family relatives. The Church also promoted the freedom of choosing with whom to marry and encouraged newly married couples to adopt neolocality habits. Polygamous marriages as well as concubinage and remarriages have been increasingly proscribed. All these policies ended up gradually imposing the nuclear family as the privileged form of family after 1500. It has therefore also contributed to the rise of the European Marriage Pattern (EMP).

The Industrial Revolution and the associated fertility transition came with a decline of the influence of the Church in individual lives, namely secularization (see Baudin, 2010; Coale & Watkins, 1986; among others). For this reason, former bans on remarriage and out-of-wedlock births, for example, became less and less stringent from the 20th century on. It implies that after a long time, family forms in Europe became more diverse again (Esping-Andersen & Billari, 2015).

In other religious and cultural contexts, the nuclear family is not necessarily the prevailing family form. For instance, Muslims, but also Buddhists, have a significant share of their population living under the polygamous—mainly polygynous—marital arrangements. These forms of marriage are strongly associated with the complex family type.16 Between 2010 and 2018, polygamy was mostly present in sub-Saharan Africa (Pew Research Center, 2019), where the complex family is also widespread (Figure 4).17 That being said, polygamy is not the only reason why people live in extended-family households in this region. Indeed, Pew Research Center (2019) shows that the proportion of persons living in polygamous households is always significantly lower than the proportion of persons living in extended-family households, even in western sub-Saharan Africa where polygamy is widespread.18


This article provided a series of arguments showing that each family form is well adapted to specific economic and institutional environments. For instance, stem and complex families are efficient in highly uncertain environments while the nuclear family is more adapted to periods of intense cultural change and rapid technological transformation or industrialization. Empirical evidence shows several types of families exist, and that the nuclear family should not be considered as the only way in which families are organized. Building a definitive model of the family is an impossible task, families are malleable objects evolving along our development paths, historical accidents, and cultural evolution.

The discussion here demonstrates that there is still a lot to learn on the functioning of families. Until now, social scientists have documented differences across family types over time and space. The functioning, or intrahousehold decision processes, of these different families is still to be better understood. This is of extreme importance to assess questions regarding how individuals across different families make decisions related to children, education, labor supply, bequests, etc. In turn this will reinforce our capacity to understand key economic phenomena like economic growth, inequalities, institutional developments, or incidence of public policies. The authors believe that in this respect it is essential to build new structural models that allow to integrate empirical findings.

Table 5. Summary Table: What Are Family Types Related to?

Family type







High education, GDP per capita, and employment

Land scarcity

High female-to-male enrollment rate ratio in upper primary schools

Collective choice model


Le Play (1884)

Le Play (1884)

Greif (2006)

Duranton et al. (2009)

Guirkinger and Platteau (2015)

Bertocchi and Bozzano (2015)

Baland and Ziparo (2018)

Schulz et al. (2019)


Old age support

Frictions in the housing and construction markets

Risk of unemployment

Agricultural economy

Impartible inheritance

Less domestic violence

Dynastic preferences

Zink (1993)

Zink (1993)

Kaplan (2012)

Pensieroso and Sommacal (2019)

Tur-Prats (2019)

Tur-Prats (2019)

Gobbi and Goñi (2021)


Traditionalism, perpetuation of customs

High income uncertainty

Poor public good provisions

Poor institutions

Poor insurance and credit markets

Pareto efficient allocation of resources

Agricultural production

Le Play (1884)

Foster and Rosenzweig (2002)

Cox and Fafchamps (2007)

Alesina and Giuliano (2015)

Baland et al. (2017)

Rangel and Thomas (2019)

Rangel and Thomas (2019)

Further Reading

  • Anderson, S., & Bidner, C. (2022a). An institutional perspective on the economics of the family. CEPR.
  • Anderson, S., & Bidner, C. (2022b). How economics can contribute to evolutionary perspectives on the family. CEPR.


  • Akresh, R., Chen, J. J., & Moore, C. T. (2012). Productive efficiency and the scope for cooperation in polygynous households. American Journal of Agricultural Economics, 94(2), 395–401.
  • Alesina, A., Algan, Y., Cahuc, P., & Giuliano, P. (2015). Family values and the regulation of labor. Journal of the European Economic Association, 13(4), 599–630.
  • Alesina, A., & Giuliano, P. (2015). Culture and institutions. Journal of Economic Literature, 53(4), 898–944.
  • Alesina, A., Giuliano, P., & Nunn, N. (2013). On the origins of gender roles: Women and the plough. The Quarterly Journal of Economics, 128(2), 469–530.
  • Alger, I., & Weibull, J. W. (2008). Family ties, incentives and development: A model of coerced altruism. SSE/EFI Working Paper Series in Economics and Finance, 681, 7–10.
  • Alston, L. J., & Schapiro, M. O. (1984). Inheritance laws across colonies: Causes and consequences. The Journal of Economic History, 44(2), 277–287.
  • Anderson, S., & Baland, J.-M. (2002). The economics of roscas and intrahousehold resource allocation. The Quarterly Journal of Economics, 117(3), 963–995.
  • Apps, P., & Rees, R. (1988). Taxation and the household. Journal of Public Economics, 35, 355–369.
  • Baland, J. M., Bonjean, I., Guirkinger, C., & Ziparo, R. (2017). The economic consequences of mutual help in extended families. Journal of Development Economics, 123, 38–56.
  • Baland, J.-M., & Ziparo, R. (2018). Chapter 4: Intra-household bargaining in poor countries. In S. Anderson, L. Beaman, & J.–P. Platteau (Eds.), Towards gender equity in development (pp. 69–96). Oxford University Press.
  • Barr, A., Dekker, M., Janssens, W., Kebede, B., & Kramer, B. (2019). Cooperation in polygynous households. American Economic Journal: Applied Economics, 11(2), 266–283.
  • Bau, N. (2021). Can policy change culture? Government pension plans and traditional kinship practices. American Economic Review, 111(6), 1880–1917.
  • Bau, N., & Fernández, R. (2022). Culture and the family. In S. Lundberg & A. Voena (Eds.), Handbook of family economics.
  • Baudin, T. (2010). A role for cultural transmission in fertility transitions. Macroeconomic Dynamics, 14(4), 454–481.
  • Becker, G. (1981). A treatise of the family. Harvard University Press.
  • Becker, G. S., & Barro, R. J. (1988). A reformulation of the economic theory of fertility. The Quarterly Journal of Economics, 103(1), 1–25.
  • Berkner, L. K., & Mendels, F. F. (1978). Inheritance systems, family structure, and demographic patterns in Western Europe, 1700–1900. In C. Tilly (Ed.), Historical studies of changing fertility (pp. 209–224). Princeton University Press.
  • Bertocchi, G., & Bozzano, M. (2015). Family structure and the education gender gap: Evidence from Italian provinces. CESifo Economic Studies, 61(1), 263–300.
  • Bezin, E., Verdier, T., & Zenou, Y. (2021). Crime, broken families, and punishment. American Economic Journal: Microeconomics, 14(4), 723–760.
  • Bisin, A., & Verdier, T. (2001). The economics of cultural transmission and the dynamics of preferences. Journal of Economic Theory, 97(2), 298–319.
  • Blundell, R., Chiappori, P.-A., & Meghir, C. (2005). Collective labor supply with children. Journal of Political Economy, 113, 1277–1306.
  • Bonfield, L. (2001). Developments in European family law. In D. I. Kertzer & M. Barbagli (Eds.), Family life in early modern times 1500–1789 (The history of the European family: Vol. 1) (pp. 87–124). Yale University Press.
  • Botticini, M., & Siow, A. (1993). Why dowries? American Economic Review, 93(4), 1385–1398.
  • Bourdieu, P. (2002). Le bal des célibataires: Crise de la société paysanne en Béarn. Seuil.
  • Boyd, R., & Richerson, P. J. (1985). Culture and the evolutionary process. University of Chicago Press.
  • Browning, M., & Chiappori, P.-A. (1998). Efficient intra-household allocations: A general characterization and empirical tests. Econometrica, 66, 1241–1278.
  • Browning, M., Chiappori, P.-A., & Weiss, Y. (2014). Economics of the family. Cambridge University Press.
  • Carta, F., De Masi, L., & Profeta, P. (2021). Family culture and childcare policies. Mimeo.
  • Cavalli-Sforza, L. L., & Feldman, M. W. (1981). Cultural transmission and evolution: A quantitative approach. Princeton University Press.
  • Chabé-Ferret, B. (2019). Adherence to cultural norms and economic incentives: Evidence from fertility timing decisions. Journal of Economic Behavior and Organization, 162, 24–48.
  • Cherchye, L., De Rock, B., & Vermeulen, F. (2012). Married with children: A collective labor supply model with detailed time use and intrahousehold expenditure information. American Economic Review, 102, 3377–3405.
  • Chiappori, P.-A. (1988). Rational household labor supply. Econometrica, 56, 63–89.
  • Chiappori, P.-A., & Mazzocco, M. (2017). Static and intertemporal household decisions. Journal of Economic Literature, 55, 985–1045.
  • Coale, A., & Watkins, S. (1986). The decline of fertility in Europe: The revised proceedings of a conference on the Princeton European Fertility Project. Princeton University Press.
  • Cox, D., & Fafchamps, M. (2007). Extended family and kinship networks: Economic insights and evolutionary directions. In T. P. Schultz & J. A. Strauss (Eds.), Handbook of development economics (pp. 3711–3784). Elsevier.
  • Croson, R., & Gneezy, U. (2009). Gender differences in preferences. Journal of Economic Literature, 47, 448–474.
  • de la Croix, D., & Mariani, F. (2015). From polygyny to serial monogamy: A unified theory of marriage institutions. Review of Economic Studies, 82(2), 565–607.
  • de la Croix, D., Schneider, E. B., & Weisdorf, J. (2019). Childlessness, celibacy and net fertility in pre-industrial England: The middle-class evolutionary advantage. Journal of Economic Growth, 24(3), 223–256.
  • De Moor, T., & van Zanden, J. L. (2010). Girl power: The European marriage pattern and labour markets in the North Sea region in the late medieval and early modern period. The Economic History Review, 63(1), 1–33.
  • De Nardi, M., French, E. B., Jones, J. B., & McGee, R. (2021). Why do couples and singles save during retirement? CEPR.
  • Doepke, M., & Tertilt, M. (2016). Families in macroeconomics. In J. B. Taylor & H. Uhlig (Eds.), Handbook of macroeconomics (Vol. 2, pp. 1790–1882). Elsevier.
  • Duranton, G., Rodríguez-Pose, A., & Sandall, R. (2009). Family types and the persistence of regional disparities in Europe. Economic Geography, 85(1), 23–47.
  • Ebenstein, A. (2014). Patrilocality and missing women.
  • Edlund, L., & Rahman, A. (2005). Household structure and child outcomes: Nuclear vs. extended families—Evidence from Bangladesh.
  • Esping-Andersen, G., & Billari, F. (2015). Re-theorizing family demographics. Population and Development Review, 41(1), 1–31.
  • Fafchamps, M. (2001). Intrahousehold access to land and sources of inefficiency: Theory and concepts. In A. de Janvry, G. Gordillo, E. Sadoulet, & J.–P. Platteau (Eds.), Access to land, rural poverty, and public action (pp. 68–96). Oxford University Press.
  • Fenske, J. (2014). Ecology, trade, and states in pre-colonial Africa. Journal of European Economic Association, 12(3), 612–640.
  • Foreman-Peck, J. (2011). The Western European marriage pattern and economic development. Explorations in Economic History, 48(2), 292–309.
  • Foster, A., & Rosenzweig, M. R. (2002). Household division and rural economic growth. Review of Economic Studies, 69(4), 839–869.
  • Galasso, V., & Profeta, P. (2018). When the state mirrors the family: The design of pension systems. Journal of the European Economic Association, 16(6), 1712–1763.
  • Galperti, S., & Strulovici, B. (2017). A theory of intergenerational altruism. Econometrica, 85(4), 1175–1218.
  • Giuliano, P., & Nunn, N. (2018). Ancestral characteristics of modern populations. Economic History of Developing Regions, 33(1), 1–17.
  • Gobbi, P. E. (2018). Childcare and commitment within households. Journal of Economic Theory, 176, 503–551.
  • Gobbi, P. E., & Goñi, M. (2021). Childless aristocrats: Inheritance and the extensive margin of fertility. The Economic Journal, 131(637), 2089–2118.
  • Goody, J., & Goody, J. R. (1983). The development of the family and marriage in Europe. Cambridge University Press.
  • Goody, J., Thirsk, J., & Thompson, E. P. (1978). Family and inheritance: Rural society in Western Europe; 1200–1800. Cambridge University Press.
  • Greenwood, J., Guner, N., & Vandenbroucke, G. (2017). Family economics writ large. Journal of Economic Literature, 55(4), 1346–1434.
  • Greif, A. (2006). Family structure, institutions, and growth: The origins and implications of western corporations. AEA Papers and Proceedings, 96(2), 308–312.
  • Guirkinger, C., & Platteau, J. P. (2015). Transformation of the family farm under rising land pressure: A theoretical essay. Journal of Comparative Economics, 43(1), 112–137.
  • Guirkinger, C., & Platteau, J. P. (2020). The dynamics of family systems. In J. M. Baland, F. Bourguignon, J. P. Platteau, & T. Verdier (Eds.), The handbook of economic development and institutions (pp. 449–519). Princeton University Press.
  • Habakkuk, H. J. (1955). Family structure and economic change in nineteenth-century Europe. The Journal of Economic History, 15(1), 1–12.
  • Hajnal, J. (1965). European marriage patterns in perspective. In D. V. Glass & D. E. C. Eversley (Eds.), Population in history (pp. 101–143). Routledge.
  • Hartung, J. (1982). Polygyny and inheritance of wealth. Current Anthropology, 23(1), 1–12.
  • Henderson, J. V., Squires, T., Storeygard, A., & Weil, D. (2018). The global distribution of economic activity: Nature, history and the role of trade. The Quarterly Journal of Economics, 133, 357–406.
  • Kaplan, G. (2012). Moving back home: Insurance against labor market risk. Journal of Political Economy, 120(3), 446–512.
  • Kinnan, C. (2021). Distinguishing barriers to insurance in Thai villages. Journal of Human Resources, 57(1), 44–78.
  • Krishnan, P., & Sciubba, E. (2009). Links and architecture in village networks. The Economic Journal, 119(537), 917–949.
  • Laslett, P. (1972). Household and family in past time. Cambridge University Press.
  • Le Bris, D. (2020). Family characteristics and economic development.
  • Le Play, F. (1884). L’organisation de la famille selon le vrai modèle signalé par l’histoire de toutes les races et de tous les temps. Mame.
  • Le Roy Ladurie, E. (1972). Système de la coutume: Structures familiales et coutumes d’héritage en France au XVIe siècle. Annales: E´conomies, Sociétés, Civilisations, 4–5, 825–846.
  • Lindbeck, A., Nyberg, S., & Weibull, J. W. (1999). Social norms and economic incentives in the welfare state. The Quarterly Journal of Economics, 114(1), 1–35.
  • Loper, J. (2021). Women’s position in ancestral societies and female HIV: The long-term effect of matrilineality in sub-Saharan Africa.
  • Lowes, S. (2018). Matrilineal kinship and spousal cooperation: Evidence from the matrilineal belt.
  • Lowes, S. (2021). Ethnographic and field data in historical economics. In A. Bisin & G. Federico (Eds.), The handbook of historical economics (pp. 147–177). Academic Press.
  • Michalopoulos, S. (2012). The origins of ethnolinguistic diversity. American Economic Review, 102, 1508–1539.
  • Murdock, G. P. (1959). Africa: Its people and their culture history. McGraw Hill.
  • Murdock, G. P. (1967). Ethnographic atlas. University of Pittsburgh Press.
  • Pensieroso, L., & Sommacal, A. (2014). Economic development and the family structure: From pater familias to the nuclear family. European Economic Review, 71, 80–100.
  • Pensieroso, L., & Sommacal, A. (2019). Agriculture to industry: The end of intergenerational coresidence. Review of Economic Dynamics, 34, 87–102.
  • Peters, A., Piazzesi, M., & Schneider, M. (2021). Houses and families across countries. In Women in macro presentation.
  • Pew Research Center. (2019). Religion and living arrangements around the world.
  • Rangel, M. A., & Thomas, D. (2019). Decision-making in complex households.
  • Razin, A., & Ben-Zion, U. (1975). An intergenerational model of population growth. American Economic Review, 65(5), 923–933.
  • Reher, D. S. (1998). Family ties in Western Europe: Persistent contrasts. Population and Development Review, 24(2), 203–234.
  • Rossi, P. (2019). Strategic choices in polygamous households: Theory and evidence from Senegal. Review of Economic Studies, 86(3), 1332–1370.
  • Schulz, J. F., Bahrami-Rad, D., Beauchamp, J. P., & Henrich, J. (2019). The church, intensive kinship, and global psychological variation. Science, 366(6466), 1–12.
  • Tène, E. (2021). On the historical roots of gender norms: Evidence from matrilineal societies in sub-Saharan Africa.
  • Tertilt, M. (2005). Polygyny, fertility, and savings. Journal of Political Economy, 113(6), 1341–1371.
  • Todd, E. (1990). L’Invention de l’Europe. Seuil.
  • Todd, E. (2011). L’origine des systèmes familiaux. Seuil.
  • Tur-Prats, A. (2019). Family types and intimate partner violence: A historical perspective. The Review of Economics and Statistics, 101(5), 878–891.
  • Udry, C. (1996). Gender, agricultural production, and the theory of the household. Journal of Political Economy, 104, 1010–1046.
  • Voigtländer, N., & Voth, J. H. (2013). How the west “invented” fertility restriction. American Economic Review, 103(6), 2227–2264.
  • Zink, A. (1993). L’héritier de la maison: Géographie coutumière du Sud-Ouest de la France sous l’Ancien Régime. EHESS.


  • 1. See Becker (1981) for a first review and volume 81, issue 1 of Journal of Demographic Economics in honor of Gary Becker. See Browning et al. (2014), Doepke and Tertilt (2016), and Greenwood et al. (2017) for recent reviews of micro- and macroeconomic models.

  • 2. This article focuses on within-households arrangements and not on the links between members of different households that share a same lineage or kinship group, sometimes also referred to as extended families in the literature (Cox & Fafchamps, 2007).

  • 3. This article does not review the literature studying the emergence and effects of matrilineal versus patrilineal societies. This would add another cross-classification in terms of family types. Readers interested in this topic should read the works of Lowes (2018), Loper (2021), and Tène (2021).

  • 4. The authors of this article are of course not the first to make this point; see, for instance, Baland and Ziparo (2018), Guirkinger and Platteau (2020), and Bau and Fernández (2022) for interesting discussions.

  • 5. This article focuses on the nuclear, stem, and complex family types classification here. For research on the effects of matrilocal and patrilocal orientations, see Bau (2021).

  • 6. The same appears to be the case for historical Europe, as shown in Figure 2 of Duranton et al. (2009).

  • 7. This article distinguishes between nuclear, polygynous, or stem and complex families. There were very few stem family types and therefore they were merged with complex ones.

  • 8. This classification precedes that of Todd (2011), which is the one in Table 1. Notice that Todd (2011) highlights several corrections from his previous classification in Todd (1990) from Duranton et al. (2009).

  • 9. Following Todd’s (1990) classification, Duranton et al. (2009) distinguish between two types of nuclear families, the “egalitarian” and the “absolute” nuclear families. They find that the absolute one is better in terms of economic outcomes.

  • 10. The concept of cultural transmission proposed by Bisin and Verdier (2001) grounds on Cavalli-Sforza and Feldman (1981) and Boyd and Richerson (1985).

  • 11. Bezin et al. (2021) study the influence of family structure on cultural dynamics but not by extending Bisin and Verdier’s framework to stem and complex families. Instead, they distinguish nuclear families where the father is present from those where they are not. They show that severe crime repression tend to break nuclear families by jailing fathers, which can translate in higher probabilities for sons to commit crime in the future. Severe crime repression can then be inefficient in the long run.

  • 12. de la Croix et al. (2019) also mention high child mortality and high childlessness rates (10%–15%).

  • 13. This institution has also been linked to economic development by providing individuals with longer time horizons for human capital investment through late marriages (De Moor & van Zanden, 2010; Foreman-Peck, 2011).

  • 14. There is a large literature on temporary coresidence, which the authors did not aim to review in this article.

  • 15. See also Le Bris (2020) for an instructive discussion on the links (and missing links) between family types, religion, and institutions.

  • 16. Polygyny is a marriage between a man and two or more wives at one moment of time.

  • 17. de la Croix and Mariani (2015) propose a unified theory of the evolution of marital institutions where monogamy and serial monogamy are institutional equilibrium succeeding polygamy when income distribution becomes less unequal.

  • 18. For more studies on polygyny, see Hartung (1982), Botticini and Siow (1993), Tertilt (2005), Akresh et al. (2012), Barr et al. (2019), and Rossi (2019). Hartung (1982) and Botticini and Siow (1993) highlight the association between polygyny and the use of a bride price. Tertilt (2005) shows that polygyny harms development because men invest in wives rather than in physical assets, hence harming capital accumulation. Rossi (2019) shows that children are strategic complements within polygynous households. Inside such forms of family, women have an incentive to have children because an extra child allows women to increase their share of family resources, controlled by the man. Barr et al. (2019) show that polygynous families are less cooperative and less altruistic than monogamous families. Using data for Burkina Faso, Akresh et al. (2012) argue that selfish preferences can encourage cooperation when it comes to farm yields.