> For the complete documentation index, see [llms.txt](https://laurence-wilse-samson.gitbook.io/textbooks/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://laurence-wilse-samson.gitbook.io/textbooks/institutional-microeconomics/part-i-scarcity-property-and-exchange/ch05_nonmarket_allocation.md).

# Chapter 5 — Non-Market Allocation: Households, Reciprocity, and Culture

***

## The Kula Ring

In *Argonauts of the Western Pacific*, the anthropologist Bronislaw Malinowski (1922) described the Kula — a vast system of ceremonial exchange among the island communities of the Massim region, off the eastern tip of New Guinea. In the Kula, two kinds of objects circulated in opposite directions around a ring of islands: necklaces of red shell (*soulava*) traveled clockwise, and armbands of white shell (*mwali*) traveled counterclockwise. Each man received these objects from his Kula partners, held them for a time, and eventually passed them on to other partners.

The objects had almost no practical use: they were worn, if at all, only on ceremonial occasions, and they could not be eaten or bartered. Their value was social. Holding a famous necklace or armband conferred prestige, cemented alliances, and created obligations that stretched across hundreds of miles of open ocean. The Kula was not a market — there were no prices and no haggling — yet alongside it a practical trade in yams, fish, pottery, and canoes flourished among partners whom the ceremonial exchange had made trustworthy.

Marcel Mauss (1925), building on Malinowski's account, identified the Kula as one instance of a universal phenomenon: the *gift economy*, in which goods circulate through obligation, reciprocity, and social debt rather than purchase. In a gift economy the value of an exchange lies less in the thing exchanged than in the relationship it creates and sustains. To give is to claim standing; to receive is to accept obligation; to reciprocate is to restore equality. The gift, Mauss argued, is never free: it is a "total" social fact, at once economic, legal, moral, and religious.

This chapter departs from the usual textbook sequence, deliberately. Most textbooks introduce the price mechanism first and treat non-market allocation as a deviation — a "market failure" or a pre-modern curiosity. Chapters 2 and 4 showed that choice, exchange, and gains from trade do not require prices: bargaining in the Edgeworth box produced a contract curve with no auctioneer in sight. Before Chapter 6 introduces the price mechanism, we examine the allocative institutions that preceded markets and still govern much of daily life in the most marketized economies. The Kula Ring is not a failed market, and the household is not an imperfect firm. Both solve genuine economic problems — coordination, insurance, motivation, distribution — by mechanisms older than the price system.

## The Spine

**What problem exists?** Societies must allocate scarce resources — food, labor, care, risk-bearing — and markets are only one way to do it; for most of human history they were not the main one.

**What institution addresses it?** Households, kinship networks, gift economies, and communities allocate through reciprocity, authority, and social norms rather than prices.

**What role do prices play?** Little or none: allocation follows obligation, status, and custom. Yet these systems solve real economic problems — insurance, coordination, motivation — that markets handle differently, and not always better; and where prices are introduced, they can change what an obligation means (§5.4).

**Who captures the rents?** Living and producing together creates a surplus over what each member could secure alone — a rent in the strict sense, a return above the next-best alternative. Its division follows fallback positions that law and custom construct (property, inheritance, and divorce rules) and norms that assign roles. Historically, those with social authority — household heads, elders, patrons in gift networks — have claimed disproportionate shares, enforced by obligation and the threat of exclusion rather than by contract.

***

## 5.1 Markets Are One Allocative Mechanism

### The Primacy of Non-Market Allocation

The standard curriculum begins with markets and treats non-market allocation as an afterthought. This gets the history backwards. For most of human existence, allocation ran through kinship obligations, communal sharing, hierarchical redistribution, reciprocal exchange, and custom. Markets in the modern sense — anonymous, price-mediated, open to all — are a comparatively recent institution, and even in the most market-oriented economies much economic activity happens outside the price system.

Karl Polanyi (1944) drew the map this chapter follows. He distinguished *reciprocity* (obligatory giving among kin and partners), *redistribution* (collection to a center and allocation from it), and *householding* (production for the group's own use) from *exchange* at market prices, and argued that before the nineteenth century the economy was "embedded" in social relations governed by the first three. This section treats reciprocity and redistribution, §5.2–§5.4 the household, and Chapter 6 the market.

Consider a household in a rich country. Parents do not charge children for meals; spouses do not invoice each other for chores. Who cooks, who drives the children to school, and who manages the money is settled by habit, negotiation, and norms. The household is a small planned economy — the allocative institution most people deal with most of the time.

Beyond the household, communities allocate through shared labor (barn-raising, communal rice transplanting), mutual aid (rotating savings groups, informal insurance), and cultural institutions (tithing, feasts, potlatch ceremonies). Even within market economies, firms allocate internally by hierarchy rather than by price (Chapter 11).

### Gift Exchange and Reciprocity

Mauss's (1925) *The Gift* identified three obligations that structure non-market exchange across cultures: to give, to receive, and to reciprocate. Gift exchange is not charity; it is a system of mutual obligation that creates and sustains social bonds. The gift creates a debt; the return gift discharges it while creating a new one; the cycle continues, binding the parties in an ongoing relationship.

The same structure is visible today. Business gifts, dinner invitations, academic co-authorship, and political favors all follow the logic of reciprocity. The obligation is not contractual — nothing specifies what must be returned, or when — but it binds. Failing to reciprocate, or reciprocating meanly, invites sanctions (gossip, exclusion, loss of standing) that can be as effective as a legal penalty. Competitive giving is not a violation of the system but one of its forms: in the potlatch of the Pacific Northwest, chiefs claimed rank by giving away, and sometimes destroying, wealth on a scale their rivals could not match — Mauss's point that to give is also to claim superiority.

Reciprocity also sustains cooperation where contracts cannot be enforced. Fafchamps (2004) documents how trade in sub-Saharan Africa runs on relationships: traders and manufacturers extend credit and share information about unreliable partners because the prospect of future dealings makes cooperation self-enforcing — the logic of the Maghribi coalition (§1.6). Such mechanisms substitute for formal institutions where these are weak and complement them where they are strong. (Appendix 5, §A5.4–§A5.5, formalizes reciprocity both as a preference and as a self-enforcing arrangement.)

### Redistributive Economies and Moral Economies

Some societies allocate primarily through centralized redistribution. A chief, temple, or state collects tribute and redistributes it according to political, religious, or social criteria rather than willingness to pay. The temple economies of ancient Mesopotamia, the redistribution systems of Polynesian chiefdoms, and the welfare states of modern Scandinavia all exhibit this logic, at very different scales and with very different institutional forms.

E. P. Thompson (1971) gave the old phrase *moral economy* its modern meaning: the customary norms — a just price for bread, a duty not to profit from dearth — that eighteenth-century English crowds enforced through food riots when dealers and millers broke them. James Scott (1976) carried the concept to the peasant societies of Southeast Asia, where landlords were expected to reduce rents in bad harvest years, wealthier households to share with poorer ones, and village institutions to uphold a rough "right to subsistence". These norms were not mere charity; they were the infrastructure of risk-sharing in societies without insurance markets.

Why should protection against disaster take the form of custom rather than contract? Partly because the hazards a peasant household faces are not the kind an insurer can price. Scott's cultivators followed a "safety-first" principle, preferring arrangements that minimized the chance of ruin to those that maximized average income, and the hazards they faced — harvest failure, illness, a landlord's demands, war — could neither be assigned probabilities nor, often, be listed in advance. That is Knightian uncertainty (see §17.2), the natural habitat of reciprocity. An insurance contract must name and price the events it covers; an obligation to help kin and neighbors "in need" names nothing in advance, and so covers contingencies no contract could enumerate. Its weakness is enforcement: the fortunate must be willing to give. Village economies therefore insure well but not fully. Townsend (1994) finds that household consumption in Indian villages tracks village-wide consumption far more closely than the household's own income, yet full insurance is statistically rejected — as the theory of self-enforcing insurance predicts (Appendix 5, §A5.5).

When colonial states imposed market institutions — fixed cash rents in place of shares that fell with the harvest, head taxes fixed in cash, export crops in place of subsistence production — they disrupted the moral economy without replacing its insurance, and Scott traced the peasant rebellions of the early 1930s in Burma and Vietnam to exactly this squeeze. Market institutions do not simply add to an existing repertoire: they can destroy the non-market institutions they displace, and the net welfare effect can be negative.

***

## 5.2 The Household as Allocative Institution

### Becker's Household Production Model

Gary Becker (1965, 1981) treated the household not as a unit of consumption but as a unit of *production*: it combines market goods and time to produce the "commodities" that yield utility — meals (food and cooking time), a clean home (supplies and labor), child-rearing (goods and intensive time).

The full cost of any household activity is therefore its money cost plus the opportunity cost of the time it absorbs: a home-cooked meal costs the ingredients plus the value of the cook's time. As wages rise, home production becomes dearer relative to market substitutes — prepared food, restaurants, paid childcare, cleaning services — and households shift toward them. This helps explain the long twentieth-century shift from household to market production that accompanied rising women's wages; cheaper and better household appliances, which cut the time housework required, mattered too (Greenwood, Seshadri, and Yorukoglu 2005).

Formally, the household maximizes utility over commodities $$Z\_i = f\_i(x\_i, t\_i)$$, where $$x\_i$$ are market goods and $$t\_i$$ is time devoted to producing commodity $$i$$, subject to

$$
\sum\_i p\_i x\_i \leq wH + y \qquad \text{(budget)}
$$

$$
\sum\_i t\_i + H \leq T \qquad \text{(time)}
$$

where $$w$$ is the wage, $$H$$ is hours of market work, $$y$$ is non-labor income, and $$T$$ is total time. Substituting $$H = T - \sum\_i t\_i$$ into the budget gives the *full income* constraint:

$$
\sum\_i (p\_i x\_i + w t\_i) \leq wT + y
$$

The term $$wt\_i$$ is the time cost of activity $$i$$, the earnings forgone by spending time at home. Full income $$wT + y$$ is what the household could earn by selling all its time, and it is spent on commodities whose *full price* includes goods and time; when each unit of $$Z\_i$$ requires fixed amounts of goods and time — Becker's own assumption — that full price is a constant (Appendix 5, §A5.1). Two cautions. Valuing time at the wage is right only for someone who does some paid work: a full-time homemaker's hour may be worth more at home. And the model has a single decision-maker, an assumption the rest of this section takes apart.

### Time Allocation and the Value of Domestic Work

Becker's framework exposes an asymmetry in measurement. National income accounts (Chapter 2) exclude household production: when a parent stays home with a child, GDP is unchanged; when the same parent hires a nanny and takes a paid job, GDP rises by the nanny's wages and the parent's earnings, though the amount of childcare is the same.

The omitted production is large. Bridgman et al. (2012), at the US Bureau of Economic Analysis, estimate that counting household production would raise measured US GDP by about 26 percent in 2010, and by more in 1965, when more hours went into home production. Where more production happens inside the household — fetching water, gathering fuel, processing food — the omission matters even more, and because women do most unpaid work everywhere, it systematically understates their contribution.

Time-use surveys, in which respondents record what they did hour by hour, are the main window onto household production. American time-diary studies show women's weekly hours of housework falling by almost half between 1965 and 1995 while men's roughly doubled from a much lower base — the gap narrowed mainly because women did less (Bianchi et al. 2000). The American Time Use Survey, run by the Bureau of Labor Statistics since 2003, shows that a gap remains even in couples where both partners work full time, and it does not close when the wife out-earns her husband: Bertrand, Kamenica, and Pan (2015) find that the gender gap in home production is, if anything, larger in such couples — a pattern no comparative-advantage story predicts, pointing instead to the identity norms and equilibria of §5.4.

Across countries the direction of the gap never changes; its size does. Women do more unpaid work than men in every country with comparable time-use data (Miranda 2011), and worldwide they perform about three-quarters of all unpaid care work — around 4.4 hours a day against 1.4 hours for men (Addati et al. 2018). Measured on a common OECD definition, women's unpaid work is about 1.3 times men's in Sweden and 1.8 times in the United Kingdom, but 5.5 times in Japan and nearly 7 times in India in 1998–99 (Figure 5.1). A ratio that differs threefold between Britain and Japan, two rich economies with similar technology, is hard to attribute to biology or comparative advantage alone; it points to institutions — care services, leave policies, norms — which is where the rest of the chapter looks.

![The Gender Gap in Unpaid Work](https://2171264621-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2FgdyZSdo6Lnq7kmkrkZqF%2Fuploads%2Fgit-blob-a94dd000cbfcabc135c098c233d9d56551f6478f%2Ffig05_gender_unpaid_work.png?alt=media)

*Figure 5.1: The Gender Gap in Unpaid Work. Minutes per day of unpaid work (housework, shopping, care, volunteering, and related travel) by women and men in four countries, with women's time as a multiple of men's. Women do more everywhere, but the ratio runs from about 1.3 in Sweden to nearly 7 in India. Source: OECD Time Use Database, from national time-use surveys (Sweden 2010, United Kingdom 2014–15, Japan 2016, India 1998–99).*

### From Unitary to Collective: Chiappori's Model

Becker's model treats the household as a single decision-maker — a *unitary* model in which the household maximizes one utility function, as if its members shared preferences or deferred to a benevolent head. This is convenient but empirically wrong. Members differ in preferences, outside options, and bargaining power, and the household's choices reflect an internal negotiation rather than a common objective.

Pierre-André Chiappori (1988, 1992) developed the *collective model*, which treats household decisions as the outcome of a process among members with distinct preferences and assumes only that the outcome is *Pareto efficient*: no reallocation could make one member better off without making another worse off. That is weaker than the unitary assumption, since it allows disagreement, but it is still an assumption, and we test it below. Efficiency says nothing about fairness. The jungle of §1.6 was Pareto efficient too; the collective model places the household on its frontier and leaves the question of *where* on the frontier — who gets what — to power.

If allocations are efficient and each member cares only about his or her own consumption and leisure, the household behaves as if it first divided its non-labor income by a *sharing rule* $$\phi$$ and then let each member choose freely given his or her share. The sharing rule depends on wages, non-labor income, and *distribution factors* $$z$$: variables that affect bargaining power but enter neither preferences nor the budget constraint, such as the sex ratio in the local marriage market, divorce and property law, and the share of non-labor income each spouse controls. Wages are not distribution factors, because they change the budget itself. With household public goods such as children's welfare, the decentralization works conditional on the level of those goods (§5.7; Appendix 5, §A5.2).

The distinction matters for policy. Under the unitary model it does not matter which member receives a cash transfer; under the collective model it can, because the recipient's identity can move the sharing rule.

### Testing the Collective Model

Formally, with a household public good $$G$$ (children's welfare, the quality of the home) bought at price $$p\_G$$, efficient allocations solve

$$
\max\_{x\_A, x\_B, G} ; \lambda(z), U\_A(x\_A, G) + \[1 - \lambda(z)], U\_B(x\_B, G) \quad \text{s.t. } x\_A + x\_B + p\_G G = m
$$

where $$m$$ is household income and $$\lambda(z) \in (0,1)$$ is the **Pareto weight** on A — not a preference parameter but an institutional outcome, shaped by law, norms, and fallback positions. Change a distribution factor, such as divorce law or the local sex ratio, and $$\lambda$$ moves, and with it every dimension of the allocation, although income and preferences are unchanged.

Because distribution factors move the allocation only through that single channel, they impose a testable restriction: if two of them — say the sex ratio and divorce law — work only by shifting the Pareto weight, the ratio of their effects must be the same for every outcome, whether A's labor supply, B's, or spending on any good (Bourguignon, Browning, and Chiappori 2009; §5.7). The unitary model has no such restriction, because in it distribution factors have no effect at all. Browning and Chiappori (1998) could not reject the corresponding restriction on Canadian couples' demands, although those demands violated the symmetry a unitary household would satisfy; Chiappori, Fortin, and Lacroix (2002) found US state divorce laws and sex ratios shifting husbands' and wives' labor supply as the model predicts, with its cross-equation restrictions not rejected. From such responses the sharing rule itself can be recovered, up to an additive constant (Chiappori 1992).

Is the household efficient, then? Tests like these fail to reject efficiency in many data sets, but not everywhere — and the failures are instructive.

> **How We Know: Udry (1996) — Testing Household Efficiency Plot by Plot**
>
> An efficient household must at least farm efficiently: labor and fertilizer should go wherever they yield most, whichever spouse controls the plot. Udry (1996) tests this with detailed agricultural survey data from Burkina Faso, where husbands and wives cultivate separate plots. The design compares plots within the same household, planted with the same crop in the same year — a household-by-crop-by-year fixed effect — so prices, weather, household wealth, and shared preferences cannot drive the comparison, while plot size and soil characteristics are controlled directly. Under efficiency, the gender of a plot's controller should not predict its yield. It does: plots controlled by women yield significantly less, because they receive much less labor and fertilizer per hectare than comparable plots controlled by men, and reallocating inputs within the household would raise its output by about 6 percent. The main threat is that women's plots are worse in ways the survey does not record, in which case giving them fewer inputs could be efficient; that is why the detailed plot controls matter. Efficiency — the collective model's core assumption — fails here, and the failure points to the institutions that govern who controls land and labor.

In Côte d'Ivoire, rainfall shocks to crops traditionally grown by women shift household spending differently from shocks to men's crops: spending follows the source of income, contrary to income pooling and to full insurance within the household (Duflo and Udry 2004). In rural Ethiopia, adults' nutrition does not move together as full risk-sharing requires; in poorer households in the south, women bear more of the adverse shocks (Dercon and Krishnan 2000). Efficiency is a hypothesis, not an axiom, and its failures are informative about the institutions — land rights, separate spheres of responsibility, the terms of exit — that govern the household.

***

## 5.3 Intra-Household Bargaining

### Nash Bargaining Within the Household

The bargaining approach applies the Nash bargaining solution of Chapter 4 (formal treatment in Appendix 4) to the household. Two spouses maximize the Nash product

$$
\max\_{(u\_A, u\_B)} (u\_A - d\_A)^{\beta}(u\_B - d\_B)^{1-\beta}
$$

over feasible utility pairs, where $$u\_A$$ and $$u\_B$$ are the spouses' utilities from the household allocation, $$d\_A$$ and $$d\_B$$ their disagreement payoffs, and $$\beta$$ indexes bargaining power. A spouse with a better disagreement payoff — higher earning potential, favorable divorce and property law, a social safety net, a supportive natal family — commands a larger share of household resources. Bargaining is one of the efficient processes the collective model of §5.2 allows, and it gives the sharing rule a concrete form.

### Worked Example: The Sharing Rule in Practice

Consider a household with two members, A and B, sharing total consumption of 100. Each cares only about his or her own consumption: $$U\_A(x\_A) = \ln x\_A$$ and $$U\_B(x\_B) = \ln x\_B$$, with $$x\_A + x\_B = 100$$.

Under a **unitary model**, the household maximizes some aggregate utility — say, $$\ln x\_A + \ln x\_B$$ — which yields the equal split $$x\_A = x\_B = 50$$, regardless of who earned the income.

Under the **collective model**, a sharing rule gives A the amount $$\phi$$ and B the amount $$100 - \phi$$. Suppose the division is settled by symmetric Nash bargaining ($$\beta = 1/2$$), with fallback consumption of 35 for A and 20 for B — what each could secure after a divorce, reflecting A's higher earning power. The disagreement utilities are $$\ln 35$$ and $$\ln 20$$, and $$\phi$$ maximizes the Nash product

$$
(\ln \phi - \ln 35)(\ln(100 - \phi) - \ln 20).
$$

Solving numerically gives $$\phi \approx 58.6$$: A gets more because her fallback is better.

Now suppose a divorce-law reform raises A's fallback consumption to 45. The solution moves to $$\phi \approx 63.5$$: A commands a larger share *within the marriage*, although no one divorces. Figure 5.2 shows the Nash solution sliding along the frontier toward A and the sharing rule rising steadily with A's fallback, while the unitary model predicts 50 whatever happens to it — it cannot even represent the question.

![Bargaining over the Household Frontier](https://2171264621-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2FgdyZSdo6Lnq7kmkrkZqF%2Fuploads%2Fgit-blob-5889c35f7379ddf59d0da3cda369eba71f321644%2Ffig05_household_nash_bargaining.png?alt=media)

*Figure 5.2: Bargaining over the Household Frontier. Panel (a) plots the frontier on logarithmic scales, so distances are utility differences, with the Nash solution before and after A's fallback rises from 35 to 45 (dashed curves: Nash-product level sets tangent to the frontier). Panel (b) shows A's share rising with her fallback; the unitary model predicts 50 throughout. Source: Author's illustration of the worked example in §5.3.*

### What Is the Disagreement Point?

The first household bargaining models took the disagreement point to be divorce (Manser and Brown 1980; McElroy and Horney 1981): each spouse's threat is to leave, so anything that makes leaving less costly raises that spouse's share. Lundberg and Pollak (1993) objected that couples rarely bargain over everyday matters in the shadow of divorce. In their *separate spheres* model the threat point is a non-cooperative marriage, in which each spouse falls back on the tasks and spending that convention assigns to his or her sphere; divorce is an outside option that constrains the bargain only when it beats that — the household version of the outside option principle (Chapter 4; Appendix 4, §A4.4). Changing the terms of divorce then shifts allocations mainly in marriages close to breaking up, while changing what each spouse commands inside the marriage — who receives a child benefit, say — shifts them everywhere. The UK child-benefit evidence below is consistent with the second half of that prediction.

Fallbacks are not the whole story. Amartya Sen (1990) called household relations "cooperative conflicts": cooperation benefits both parties but its terms divide them, and the terms depend also on perceptions — of each member's contribution and of whose interests count. Unpaid work that no one measures is easily perceived as no contribution at all, so measurement (Figure 5.1) is itself a bargaining issue. Agarwal (1997) adds that norms both limit what can be bargained over and are themselves bargained over, in the household and beyond it.

### How Institutions Shift Bargaining Power

**Marriage markets.** Fallbacks are set partly before a marriage begins. In Becker's (1973) theory of marriage, the terms on which partners match and divide the gains from marriage depend on the alternatives the marriage market offers each of them — which is why the sex ratio is a distribution factor. Angrist (2002) exploits the large differences in sex ratios across immigrant groups in early twentieth-century America, which mattered because the children of immigrants mostly married within their own group: second-generation women in groups where men were relatively plentiful were more likely to marry and less likely to work for pay. Anticipated shares can also shape investment before marriage. Chiappori, Iyigun, and Weiss (2009) show how a marriage-market return adds to the return to schooling, and argue that it helps explain why women's college attendance overtook men's.

**Divorce law.** When divorce becomes easier, the fallback of the spouse who would leave improves, shifting bargaining power toward that spouse *within the marriage*. Exploiting the staggered adoption of unilateral divorce across US states, which let either spouse end a marriage without the other's consent, Stevenson and Wolfers (2006) find that it was followed by declines in domestic violence and female suicide — consistent with stronger bargaining positions for women in marriages that did not end. Effects this broad suggest that, for the most consequential matters, divorce is a credible threat in many marriages, not only those near the edge; the divorce-threat and separate-spheres models are best read as describing different decisions within the same household.

**Property rights.** Fallback positions are legal constructions: the property regime determines what each spouse can claim if the marriage ends, and so what each brings to the table while it lasts. Under *community property* (much of continental Europe and Latin America, several US states), assets acquired during the marriage are owned jointly; under the traditional *common-law* rule of England and most US states, they belonged to whoever held title — historically, the husband. American common-law states moved in the 1970s and 1980s to "equitable distribution", which lets courts divide marital property regardless of title and was meant to recognize the homemaker's contribution.

In much of South Asia and sub-Saharan Africa, inheritance law or custom has long limited women's claims on land. Deininger, Goyal, and Nagarajan (2013) study the amendments to the Hindu Succession Act, adopted by several Indian states before a national amendment in 2005, that gave daughters the same rights as sons in jointly held family property. Exploiting the staggered state reforms, they find that daughters became more likely to inherit land — though large gaps with sons remained — and that girls' schooling rose. Later work on the same reforms finds gains in women's autonomy within the household and in their labor supply (Heath and Tan 2020).

Insecure rights also waste resources. Goldstein and Udry (2008) study food-crop plots in Akwapim, in southern Ghana, where leaving land fallow is the main investment in soil fertility. Because fallow land risks being reallocated to someone else, farmers without positions in the local political hierarchy — women above all — fallowed their plots for much shorter periods, and these shorter fallows account for the entire difference between profits on husbands' and wives' plots: insecure tenure, not ignorance, lowered the return to investing in the land. Ownership shapes spending too. Doss (2006), using household survey data from Ghana, finds that the share of household assets — farmland above all — owned by women is associated with different budget shares for household spending, holding total spending constant: who owns the assets matters for how the money is spent, which the collective model allows and the unitary model forbids.

Financial inclusion works the same way; in many Western countries married women could not borrow or open bank accounts without a husband's consent until the 1960s or 1970s. In a randomized experiment in Niger, Aker et al. (2016) delivered a cash transfer to women either in cash or through mobile phones; the mobile arm raised households' diet diversity, which the authors attribute partly to time saved in collecting the transfer and partly to changes in intrahousehold decision-making.

**Employment opportunities.** As women's earnings rose over the twentieth century, their fallback positions improved; Goldin (2006) calls the transformation of women's employment, education, and expectations since the 1970s a "quiet revolution".

**Social safety nets.** Programs that direct resources to specific members — child allowances paid to mothers, pensions paid to grandmothers — shift bargaining power toward the recipient, the mechanism behind the Duflo (2003) finding in §5.6.

### Why Bargaining Power Matters for Children

A consistent finding across many studies is that when women control a larger share of household resources, more is spent on children — on food, health care, and education. Thomas (1990) finds in Brazilian household data that unearned income in the mother's hands has a much larger effect on child health and survival than the same income in the father's. Lundberg, Pollak, and Wales (1997) study a late-1970s UK reform that replaced a child tax allowance, which typically raised the father's take-home pay, with a child benefit paid in cash to the mother. Household income was essentially unchanged, yet spending shifted toward women's and children's clothing relative to men's — a before-and-after comparison that other changes of the period could in principle confound, but whose specific pattern is hard to explain without a change in who controlled the money.

For child welfare, then, it may matter not only *how much* a household receives but *who* receives it. How far the logic generalizes is another question. Duflo (2012), surveying the literature, finds that the links run both ways — economic development reduces inequality between men and women, and empowering women can benefit children and development — but concludes that each link is probably too weak to set off a self-sustaining virtuous circle, so that continuing policy commitment to equality for its own sake may be needed.

The household bargain also shapes labor markets. A woman whose domestic responsibilities are fixed by norm faces a higher reservation wage for market work (she must earn enough to replace her domestic labor) and a narrower set of feasible occupations (she needs flexibility many jobs do not offer) — a thread we take up in Chapter 13 (§13.2 on reservation wages and labor supply, §13.6 on flexibility and the child penalty).

***

## 5.4 Coordination Within Households and Communities

### Coordination Problems: Who Does What?

Many household decisions are coordination problems rather than bargaining problems. Both parents want the child collected from school, neither much cares who does it, and both prefer that exactly one of them goes. In the language of game theory this is an *anti-coordination* (or "who-does-it") game, since the players must choose *different* actions; but it poses a coordination problem all the same, because the two must settle on *which* of the two asymmetric outcomes to play.

The game has two pure-strategy Nash equilibria — A goes and B does not, or the reverse — and a mixed-strategy equilibrium in which each goes with some probability (the formal definitions are collected in the Game Theory Toolkit, Appendix 0A–0C). The mixed equilibrium produces miscoordination (both go, or neither) with positive probability, so when neither parent cares who goes, both are worse off in it than in either pure equilibrium. The problem is *selecting* an equilibrium, which is where norms, custom, and authority come in.

### When Prices Crowd Out Norms

Norms and prices do not simply add up. Titmuss (1970) conjectured that paying blood donors would reduce the supply and quality of donated blood by turning a gift into a sale. The sharpest evidence of such crowding out involves the pick-up problem itself. Gneezy and Rustichini (2000) studied Israeli day-care centers that introduced a small fine for parents who collected their children late. Late pick-ups *increased*, and stayed high after the fine was withdrawn: the fine had turned a moral obligation to the teachers into a price, and a cheap one. Bowles (2016) collects many such cases and draws the general lesson: incentives act on people who also have social motives, and they can change what a situation means to them.

### Gender Norms as Coordination Equilibria

Who earns, who does housework, and who cares for children is not decided afresh in each household. It follows patterns — *gender norms*, socially shared expectations about appropriate behavior for men and women — that vary across cultures and periods but are remarkably stable within any given setting.

From the perspective of game theory, gender norms are *equilibria*: self-enforcing expectations that are individually rational given the expectation that others will follow them. If everyone expects women to do the cooking, a household in which the woman cooks and the man earns meets expectations, avoids conflict, and saves the cost of negotiating an alternative. The arrangement may be inefficient — perhaps the woman is the better earner and the man the better cook — but deviating is individually costly (disapproval, threats to identity, coordination difficulties) even when the deviation would be collectively beneficial. In the two-person version of Appendix 5, §A5.3, both the traditional and the role-reversed division of labor can be equilibria, and conformity pressure widens the range of beliefs from which the traditional one is reached (Figure A5.1).

Fernández (2013) offers a mechanism for the slow pace of change. In her model, women are uncertain about the long-run costs of market work for married women and their families, and each generation learns about them partly from the experience of the previous one. When few women work, evidence accumulates slowly and beliefs barely move; as participation rises, learning accelerates — generating the S-shaped rise in American married women's participation over the twentieth century. Culture here is a set of beliefs updated by evidence rather than a fixed taste — which helps explain why the division of household labor has adjusted far less than the division of market labor, even where women's employment has risen dramatically.

### Cultural Norms and Their Persistence

Some norms persist for centuries. Ester Boserup (1970) observed that plough agriculture, which demands upper-body strength and the control of draft animals, was done mainly by men, while hoe agriculture drew heavily on women's labor. Alesina, Giuliano, and Nunn (2013) show that descendants of plough-using societies have less equal gender norms and lower female participation in work and politics today, even after migrating to entirely different economies.

The finding challenges the common assumption that norms and institutions adjust efficiently to changing conditions: norms can persist long after their economic rationale has disappeared, so household behavior may reflect historical legacies as much as current optimization.

> **How We Know: Alesina, Giuliano, and Nunn (2013) — The Plough and Gender Norms**
>
> Alesina, Giuliano, and Nunn test Boserup's hypothesis in three steps. First, they measure each ethnic group's traditional plough use from the *Ethnographic Atlas* and map it to today's countries and individuals through ancestry: places whose ancestors farmed with the plough have lower female labor force participation, fewer women in firm ownership and in parliament, and more unequal gender attitudes, controlling for income, geography, and the characteristics of ancestral societies. Second, because plough use was itself chosen, they instrument it with the ancestral environment's suitability for crops that benefit from ploughing (wheat, barley, rye) relative to crops that benefit little (sorghum, millet). The exclusion restriction — that crop suitability affects today's norms only through the plough — is supported by controlling for overall agricultural suitability, and the IV estimates confirm the OLS pattern. Third, among the children of immigrants in the United States and Europe, who share one institutional and economic environment but differ in ancestry, the relationship persists, separating cultural transmission from the effects of current institutions and markets. A norm formed under one technology can outlive it by centuries.

***

## 5.5 From Household to Community: Cooperation and the Care Economy

### The Through-Line: Small-Group Cooperation Scales Up

The mechanisms that sustain cooperation within households — reciprocity, trust, repeated interaction, shared norms — also operate in communities, though with increasing difficulty as groups grow.

In a household of a few people, cooperation is sustained by direct reciprocity: I help you today because you helped me yesterday and because we will interact again tomorrow. In a community of dozens or hundreds, cooperation requires *indirect reciprocity* and *reputation* — the logic of the Maghribi coalition (§1.6), whose cheats were shunned by every merchant, not only the one cheated: I help you because others will see it and reciprocate later, even if you do not. At the scale of a nation, cooperation requires formal institutions — laws, taxes, enforcement — because indirect reciprocity breaks down among anonymous strangers who meet once.

This scaling logic — from household to community to nation, and on to the global commons of Chapter 27 (§27.7) — connects this chapter to the rest of the book: the informal institutions that sustain cooperation at small scales (here and in Chapter 28 on commons governance) supplement and substitute for the markets, firms, and states that dominate later chapters. The most effective real-world governance systems typically combine layers: informal norms at the base, community self-governance in the middle, and formal state institutions at the top (Ostrom 1990; see Chapter 28).

### Child-Rearing, Care Work, and the Care Economy

Child-rearing is the largest non-market productive activity in any society, and the resources devoted to it are almost entirely invisible in standard statistics.

The care economy also includes elder care, care for the disabled, and the relational labor that sustains families and communities. Folbre (2001) argues that care work is systematically undervalued and undersupplied, because it generates positive externalities — the benefits of well-raised children and well-supported elderly people accrue to society broadly, not just to the caregiver — and because it is disproportionately performed by women, whose bargaining power is weakened by the very norms that assign care to them.

The undervaluation is measurable. England, Budig, and Folbre (2002), following the same workers over time as they move between jobs, find that caring occupations — child care, teaching, and social work among them — pay less than the same workers earn in other occupations, a net "care penalty" not explained by measured skills, which the authors attribute to the devaluation of work associated with caregiving. Because women are concentrated in care occupations, the penalty adds to the gender gap in earnings.

The global scale is enormous. Oxfam (2020), valuing women's unpaid care work at the minimum wage, put it at no less than $10.8 trillion a year — a figure that depends heavily on the valuation method, but that on any method dwarfs whole industries. In low-income countries women and girls spend hours each day fetching water, gathering fuel, and processing food, time diverted from school and paid work; piped water, electricity, and better cookstoves are therefore care-economy policy as much as development policy.

Publicly funded childcare, parental leave, and elder care are likewise investments in the care economy's institutional infrastructure. The Scandinavian countries, with the most extensive public support for care, also have some of the highest rates of female employment and lowest rates of child poverty in the world — consistent with market and care production being complements when institutions support both, though the causal direction is hard to pin down.

***

## 5.6 Empirical Evidence

### Time-Use Data Across Countries

Time-use surveys are the primary empirical window onto non-market allocation, and three regularities stand out:

* **Women do more unpaid work than men in every country with comparable data**, by a factor that ranges from about 1.3 in Sweden to nearly 7 in India (1998–99; Figure 5.1).
* **In many rich countries, total work — paid plus unpaid — is roughly equal for men and women**, the "iso-work" pattern documented by Burda, Hamermesh, and Weil (2013); the gap lies in the composition, with men doing more paid work and women more unpaid work. In poorer countries, women typically work more in total.
* **The unpaid-work gap has narrowed in rich countries**, mainly because women do less housework rather than because men do much more (for the United States, Bianchi et al. 2000).

These patterns are consistent with comparative advantage (women have historically faced lower market wages), with bargaining (weaker fallback positions), and with norms — but the cross-country variation in Figure 5.1 is hard to explain with comparative advantage alone.

### Intra-Household Allocation: Duflo (2003)

Esther Duflo's (2003) study of South African pensions is among the most influential papers on the household. South Africa's Old Age Pension — a generous means-tested transfer to women aged 60 and over and men aged 65 and over — was extended in full to the Black population in the early 1990s, as apartheid was being dismantled. Duflo used the age-eligibility rules and whether the eligible household member was a grandmother or a grandfather to study the effect of the recipient's identity on children's outcomes.

The findings were striking. Pensions received by *grandmothers* substantially improved the nutritional status (weight-for-height and height-for-age) of girls in the household, with little effect on boys; pensions received by *grandfathers* had no detectable effect on either. The pattern suggests that grandmothers invested preferentially in granddaughters.

> **How We Know: Duflo (2003) — Pensions and Child Nutrition in South Africa**
>
> Duflo treats the expansion of South Africa's Old Age Pension to the Black population as a natural experiment. Eligibility depends on age — women qualify at 60, men at 65 — so whether a child lives with a pension recipient depends on the ages of the adults in the household. Households with an eligible member may differ from others in many ways, so a simple comparison could confuse the pension's effect with the effect of living with an old person. Duflo therefore exploits the fact that height-for-age reflects nutrition accumulated over a child's whole life: she compares younger children, who spent a large part of their lives under the expanded pension, with older children, in households with and without an eligible member. Fixed differences between the two kinds of household cancel out, leaving the pension's effect on children exposed during early growth. Weight-for-height, a short-run measure, provides a check, and she controls flexibly for the ages and composition of household members. The effect appears for pensions received by grandmothers but not by grandfathers, rejecting the unitary prediction that only total household income matters — with direct implications for the design of cash-transfer programs.

### Cross-Cultural Evidence on Reciprocity and Cooperation

The cross-cultural experiments of Chapter 4 bear directly on this chapter. Henrich et al. (2005) found that offers in ultimatum, dictator, and public-goods games vary dramatically across fifteen small-scale societies, and that two group-level variables predict much of the variation: the payoffs to cooperation in everyday economic life, and the degree of market integration. More market-integrated societies behaved *more* prosocially in anonymous games, contrary to the stereotype that markets erode social norms. A later study (Henrich et al. 2010) found that market integration and participation in a world religion predict fairer offers, and that community size predicts the willingness to punish unfair ones.

The interpretation is debated, but one plausible account is that dealing with strangers in markets *trains* norms of fairness, because repeated market dealings reward reliability and punish cheating. Polanyi (1944) argued that the self-regulating market tried to "disembed" the economy from social relations, and that society pushed back; the experimental evidence suggests that even market behavior never fully disembeds, since the norms that sustain markets are themselves shaped by market experience. The relationship runs both ways, not as a one-way erosion.

***

> **Common Confusions**
>
> **"The household has a single utility function."** This is the unitary model's assumption, and it is empirically rejected: who receives income, who controls assets, and what fallback positions are available all affect household allocations. The collective model (Chiappori) replaces it with the weaker assumption that household allocations are efficient — and even that assumption fails in some settings (§5.2).
>
> **"Non-market allocation is inefficient."** Reciprocity, gift exchange, household production, and community self-governance can perform well in their institutional context. Markets are efficient under specific conditions (competition, complete markets, no externalities); non-market mechanisms under different ones (small groups, repeated interaction, shared norms). The question is which arrangement performs best for a particular allocation problem, not whether markets are inherently superior.
>
> **"Unpaid work isn't economically important."** Counting household production would have raised measured US GDP by about 26 percent in 2010 (Bridgman et al. 2012). Its exclusion from national accounts is a measurement convention, not an economic reality, and it systematically understates the contribution of those — mostly women — who do the work.
>
> **"Culture is just preferences."** Culture shapes preferences, but also constraints, beliefs, and expectations about what others will do. Gender norms are equilibria sustained by expectations and social enforcement, and beliefs that change as each generation learns from the last; treating culture as fixed tastes misses the feedback between institutions, norms, and behavior.

***

## 5.7\* Formal Appendix: Households and Reciprocity

*Full derivations appear in Appendix 5.*

**Full income (Becker 1965).** With constant returns in home production, each commodity $$Z\_i$$ has a constant *full price* $$p\_i^F$$ (goods cost plus the market value of the time it absorbs), and a two-earner household maximizes $$U(Z)$$ subject to $$\sum\_i p\_i^F Z\_i \leq (w\_A + w\_B)T + y\_A + y\_B$$. Non-labor incomes enter only through their sum: the unitary model's income-pooling prediction (§A5.1).

**The collective model (Chiappori 1988, 1992).** With private goods only, efficiency is equivalent to a sharing rule: A chooses consumption and leisure subject to $$x\_A = w\_A(T - l\_A) + \phi$$ and B subject to $$x\_B = w\_B(T - l\_B) + y - \phi$$, with $$\phi = \phi(w\_A, w\_B, y, z)$$ (with a public good, conditional on its level). Because distribution factors act only through $$\phi$$, for any two of them

$$
\frac{\partial h\_A / \partial z\_1}{\partial h\_A / \partial z\_2} = \frac{\partial h\_B / \partial z\_1}{\partial h\_B / \partial z\_2},
$$

where $$h\_j$$ is member $$j$$'s labor supply. With one distribution factor and variation in non-labor income, the partial derivatives of $$\phi$$ are identified, and so $$\phi$$ itself up to an additive constant (§A5.2).

**Gender norms as equilibria.** In a two-by-two game in which specialization yields a complementarity $$\omega$$ and gender-conforming choices a social reward $$\gamma$$, the traditional and the role-reversed divisions of labor can both be equilibria. In the mixed equilibrium the husband (partner 1) chooses market work with probability $$\sigma\_1^\* = (w\_2 - v + \omega - \gamma)/(2\omega)$$, where $$v$$ is the value of home production and $$w\_2$$ the wife's wage; a larger $$\gamma$$ lowers this threshold and enlarges the set of beliefs from which the traditional equilibrium is reached (§A5.3).

**Reciprocity.** In Rabin's (1993) fairness equilibrium, mutual cooperation in a Prisoner's Dilemma with temptation payoff $$T$$ and reward $$R$$ is an equilibrium when the weight on reciprocity is at least $$2(T - R)$$; mutual defection remains one too (§A5.4). Without any taste for fairness, reciprocity can be self-enforcing in lasting relationships: gift exchange survives if the discount factor satisfies $$\delta \geq k/g$$, the cost of a gift relative to its value to the recipient, and informal insurance only as far as the value of future insurance outweighs today's temptation to renege — which is why it is typically partial (§A5.5).

***

## Summary

Markets are one allocative mechanism among several. For most of human history, and across much of contemporary life, resources have been allocated through households, kinship networks, gift exchange, community sharing, and cultural norms — the institutional environment in which market exchange is embedded.

Gift exchange (Mauss 1925) sustains social bonds through mutual obligation, and moral economies (Thompson 1971; Scott 1976) enforce norms of subsistence and risk-sharing that markets may displace without replacing. Because obligations to help "in need" name no contingencies in advance, reciprocity can insure against hazards no one could price or list — Knightian uncertainty — though only as far as the fortunate are willing to give.

The household is the most important non-market institution. Becker's household production model reveals the full cost of household activities: goods plus the opportunity cost of time. Chiappori's collective model treats household decisions as efficient outcomes among members with distinct preferences, with the allocation depending on distribution factors that shift bargaining power. The evidence rejects the unitary model — who receives income matters, and women's income often benefits children more — and even the collective model's efficiency assumption fails in some settings. Whether the fallback in household bargaining is divorce or a non-cooperative marriage determines which policies move the allocation.

Gender norms are equilibria — self-enforcing expectations that persist through cultural transmission after the conditions that produced them have changed — and prices can crowd out the norms that sustain cooperation.

Small-group cooperation — reciprocity, reputation, shared norms — scales up with increasing difficulty, from household to community self-governance to the state. The care economy is a massive flow of non-market production that standard statistics undervalue.

***

## Key Terms

* **Non-market allocation** — Distribution of resources through mechanisms other than price-mediated market exchange
* **Forms of integration** — Polanyi's classification of how economies allocate: reciprocity, redistribution, householding, and market exchange
* **Gift economy** — A system of exchange organized by obligations to give, receive, and reciprocate (Mauss)
* **Reciprocity** — The social norm and practice of returning benefits received, creating ongoing obligations
* **Moral economy** — A social order in which customary norms of subsistence and fairness constrain market logic (Thompson; Scott)
* **Safety-first** — A decision rule that puts avoiding disaster ahead of maximizing average income
* **Household production model** — Becker's framework treating the household as a producer of commodities from market goods and time
* **Full income** — Maximum possible income if all time were devoted to market work: $$wT + y$$, with $$y$$ non-labor income
* **Full price** — The total cost of a household activity, including both goods cost and the opportunity cost of time
* **Unitary model** — The assumption that the household maximizes a single utility function; empirically rejected
* **Income pooling** — The unitary prediction that household choices depend on total income, not on who receives it
* **Collective model** — Chiappori's framework treating household allocations as efficient outcomes among members with distinct preferences
* **Sharing rule** — The function describing how household non-labor income is divided among members in the collective model
* **Distribution factors** — Variables that affect bargaining power within the household but enter neither preferences nor the budget constraint (e.g., sex ratio, divorce law, control of non-labor income)
* **Pareto weight** — The weight $$\lambda(z)$$ on a member's utility in the household's efficient program, reflecting bargaining power; varies with distribution factors
* **Intra-household bargaining** — Negotiation over allocation within the household, shaped by fallback positions, norms, and law
* **Separate spheres** — Lundberg and Pollak's household bargaining model, whose threat point is a non-cooperative marriage rather than divorce
* **Cooperative conflict** — Sen's term for relations in which cooperation benefits all parties while its terms divide them
* **Coordination game** — A strategic interaction with multiple equilibria, where the challenge is equilibrium selection
* **Anti-coordination game** — A game whose pure-strategy equilibria require the players to choose different actions (e.g., who does a task)
* **Crowding out** — The weakening of intrinsic or social motives by monetary incentives
* **Gender norms** — Socially shared expectations about appropriate behavior for men and women; function as coordination equilibria
* **Care economy** — The production of care (childcare, elder care, relational labor), mostly non-market and mostly performed by women
* **Time-use survey** — A survey in which respondents record their activities throughout the day, providing data on non-market production

***

## Problems

### \[Core]

1. **Non-market allocation.** Identify three allocation mechanisms operating in your daily life that do not use prices. For each, describe (a) what resource is being allocated, (b) what mechanism governs the allocation, and (c) what would change if the allocation were done through a market instead.
2. **Household production.** A household member can earn $30/hour in the labor market. A home-cooked dinner requires $15 in ingredients and 1.5 hours of preparation. A restaurant meal of equivalent quality costs $45. (a) What is the full cost of the home-cooked dinner? (b) Should the household member cook or eat out? (c) How does your answer change if the wage is $10/hour? $50/hour? At what wage is the household indifferent? (d) Valuing the cook's time at the wage can misstate its cost in either direction. Give one reason it might overstate the cost and one reason it might understate it. (Hint: §5.2, "Two cautions.")
3. **Unitary vs. collective model.** A government transfers $200/month to a poor household. Under the unitary model, does it matter whether the transfer goes to the husband or the wife? Under the collective model? Explain the difference and describe the empirical evidence (Duflo 2003, Thomas 1990, Lundberg et al. 1997).
4. **Gender norms as coordination equilibria.** In a household with two adults, one must earn market income and one must perform childcare. Both adults prefer market work. (a) Write this as a two-player game with two pure-strategy Nash equilibria. Is it a coordination or an anti-coordination game? (b) How might a gender norm resolve the coordination problem? (c) Why might the norm persist even when both adults have equal earning potential?

### \[Extension]

5. **Chiappori's collective model.** Two household members, A and B, have utility functions $$U\_A(x\_A, l\_A)$$ and $$U\_B(x\_B, l\_B)$$, where $$x$$ is consumption and $$l$$ is leisure. Each member $$j$$ has time endowment $$T$$ and wage $$w\_j$$ and so earns $$w\_j(T - l\_j)$$; the household also has non-labor income $$y$$. Under the collective model there is a sharing rule $$\phi$$ such that A receives $$\phi$$ of the non-labor income and B receives $$y - \phi$$. (a) Write down each member's optimization problem given the sharing rule. (b) What determines $$\phi$$? (c) A divorce-law reform improves A's outside option. How does this affect $$\phi$$, each member's labor supply (assume leisure is a normal good), and the household allocation? (d) How does this differ from the unitary model's prediction?
6. **Nash bargaining in the household.** Two spouses divide household income $$m = 100$$, and each spouse's utility equals the income he or she receives. Spouse A's disagreement payoff is $$d\_A = 30$$ and spouse B's is $$d\_B = 20$$. (a) Find the symmetric Nash bargaining solution. (b) A policy reform (e.g., improved divorce law) raises $$d\_A$$ to 40. What happens to the division? (c) In (a) and (b) the Nash solution is Pareto efficient, so changing $$d\_A$$ or $$d\_B$$ only redistributes a fixed total. Now suppose that before bargaining, B can set up a home business that raises household income from 100 to 120, at a cost of 12 borne by B alone (in the same units as income) and sunk by the time the spouses bargain. Keep $$d\_A = 40$$ and $$d\_B = 20$$. (i) Will B invest? Compare total household welfare (income net of B's cost) with and without the investment. (ii) Now suppose property law lets B keep the business if the marriage ends, so that investing raises $$d\_B$$ from 20 to 30 while $$d\_A$$ stays at 40. Will B invest now? Who gains? (iii) Explain why strengthening the weaker spouse's disagreement payoff raised total welfare here, although the Nash solution is always efficient.
7. **Gift exchange and insurance.** In a community of 100 farmers, each faces an independent 10 percent chance of crop failure in a given year. A gift-exchange norm requires those with successful harvests to share with those who fail. (a) What is the expected number of farmers needing assistance? (b) How does the gift-exchange system compare to a formal insurance market? (c) Under what conditions might the gift-exchange system outperform formal insurance? (Consider enforcement costs, moral hazard, the value of the social relationship, and risks that no insurer could list in advance.) (d) No court enforces the norm: a farmer who refuses to share is simply excluded from future help. Using Appendix 5, §A5.5, explain why such a norm may provide only partial insurance, and why the amount it can sustain rises with farmers' patience.

### \[Empirical]

8. **Time-use data.** Using American Time Use Survey (ATUS) microdata from the Bureau of Labor Statistics or IPUMS-ATUS, compare the average time spent on (a) market work, (b) household production, and (c) childcare for men and women. How do these patterns vary by employment status, education, and the presence of children? Is total work time (market plus non-market) approximately equal for men and women, as the iso-work pattern (Burda, Hamermesh, and Weil 2013) suggests?
9. **Intra-household allocation.** Read Duflo (2003) on South African pensions. (a) What is the identification strategy? (b) Why does it matter that the pension is large relative to household income? (c) How do the results for grandmothers vs. grandfathers inform the debate between unitary and collective models? (d) What policy implications follow?
10. **Cultural persistence.** Read Alesina, Giuliano, and Nunn (2013) on historical plough use and contemporary gender norms. (a) What is their measure of historical plough use? (b) What contemporary outcomes do they examine? (c) What is their identification strategy for distinguishing cultural persistence from ongoing environmental effects? (d) What are the implications for the assumption that institutions adjust efficiently to economic conditions?

***

## References

Addati, Laura, Umberto Cattaneo, Valeria Esquivel, and Isabel Valarino (2018). *Care Work and Care Jobs for the Future of Decent Work*. Geneva: International Labour Office.

Agarwal, Bina (1997). "'Bargaining' and Gender Relations: Within and Beyond the Household." *Feminist Economics*, 3(1), 1--51.

Aker, Jenny C., Rachid Boumnijel, Amanda McClelland, and Niall Tierney (2016). "Payment Mechanisms and Antipoverty Programs: Evidence from a Mobile Money Cash Transfer Experiment in Niger." *Economic Development and Cultural Change*, 65(1), 1--37.

Alesina, Alberto, Paola Giuliano, and Nathan Nunn (2013). "On the Origins of Gender Roles: Women and the Plough." *Quarterly Journal of Economics*, 128(2), 469--530.

Angrist, Joshua D. (2002). "How Do Sex Ratios Affect Marriage and Labor Markets? Evidence from America's Second Generation." *Quarterly Journal of Economics*, 117(3), 997--1038.

Becker, Gary S. (1965). "A Theory of the Allocation of Time." *Economic Journal*, 75(299), 493--517.

Becker, Gary S. (1973). "A Theory of Marriage: Part I." *Journal of Political Economy*, 81(4), 813--846.

Becker, Gary S. (1981). *A Treatise on the Family*. Cambridge, MA: Harvard University Press.

Bertrand, Marianne, Emir Kamenica, and Jessica Pan (2015). "Gender Identity and Relative Income within Households." *Quarterly Journal of Economics*, 130(2), 571--614.

Bianchi, Suzanne M., Melissa A. Milkie, Liana C. Sayer, and John P. Robinson (2000). "Is Anyone Doing the Housework? Trends in the Gender Division of Household Labor." *Social Forces*, 79(1), 191--228.

Boserup, Ester (1970). *Woman's Role in Economic Development*. London: George Allen and Unwin.

Bourguignon, François, Martin Browning, and Pierre-André Chiappori (2009). "Efficient Intra-Household Allocations and Distribution Factors: Implications and Identification." *Review of Economic Studies*, 76(2), 503--528.

Bowles, Samuel (2016). *The Moral Economy: Why Good Incentives Are No Substitute for Good Citizens*. New Haven: Yale University Press.

Bridgman, Benjamin, Andrew Dugan, Mikhael Lal, Matthew Osborne, and Shaunda Villones (2012). "Accounting for Household Production in the National Accounts, 1965--2010." *Survey of Current Business*, 92(5), 23--36.

Browning, Martin, and Pierre-André Chiappori (1998). "Efficient Intra-Household Allocations: A General Characterization and Empirical Tests." *Econometrica*, 66(6), 1241--1278.

Burda, Michael, Daniel S. Hamermesh, and Philippe Weil (2013). "Total Work and Gender: Facts and Possible Explanations." *Journal of Population Economics*, 26(1), 239--261.

Chiappori, Pierre-André (1988). "Rational Household Labor Supply." *Econometrica*, 56(1), 63--90.

Chiappori, Pierre-André (1992). "Collective Labor Supply and Welfare." *Journal of Political Economy*, 100(3), 437--467.

Chiappori, Pierre-André, Bernard Fortin, and Guy Lacroix (2002). "Marriage Market, Divorce Legislation, and Household Labor Supply." *Journal of Political Economy*, 110(1), 37--72.

Chiappori, Pierre-André, Murat Iyigun, and Yoram Weiss (2009). "Investment in Schooling and the Marriage Market." *American Economic Review*, 99(5), 1689--1713.

Deininger, Klaus, Aparajita Goyal, and Hari Nagarajan (2013). "Women's Inheritance Rights and Intergenerational Transmission of Resources in India." *Journal of Human Resources*, 48(1), 114--141.

Dercon, Stefan, and Pramila Krishnan (2000). "In Sickness and in Health: Risk Sharing within Households in Rural Ethiopia." *Journal of Political Economy*, 108(4), 688--727.

Doss, Cheryl (2006). "The Effects of Intrahousehold Property Ownership on Expenditure Patterns in Ghana." *Journal of African Economies*, 15(1), 149--180.

Duflo, Esther (2003). "Grandmothers and Granddaughters: Old-Age Pensions and Intrahousehold Allocation in South Africa." *World Bank Economic Review*, 17(1), 1--25.

Duflo, Esther (2012). "Women Empowerment and Economic Development." *Journal of Economic Literature*, 50(4), 1051--1079.

Duflo, Esther, and Christopher Udry (2004). "Intrahousehold Resource Allocation in Côte d'Ivoire: Social Norms, Separate Accounts and Consumption Choices." NBER Working Paper 10498. Cambridge, MA: National Bureau of Economic Research.

England, Paula, Michelle Budig, and Nancy Folbre (2002). "Wages of Virtue: The Relative Pay of Care Work." *Social Problems*, 49(4), 455--473.

Fafchamps, Marcel (2004). *Market Institutions in Sub-Saharan Africa: Theory and Evidence*. Cambridge, MA: MIT Press.

Fernández, Raquel (2013). "Cultural Change as Learning: The Evolution of Female Labor Force Participation over a Century." *American Economic Review*, 103(1), 472--500.

Folbre, Nancy (2001). *The Invisible Heart: Economics and Family Values*. New York: New Press.

Gneezy, Uri, and Aldo Rustichini (2000). "A Fine Is a Price." *Journal of Legal Studies*, 29(1), 1--17.

Goldin, Claudia (2006). "The Quiet Revolution That Transformed Women's Employment, Education, and Family." *American Economic Review*, 96(2), 1--21.

Goldstein, Markus, and Christopher Udry (2008). "The Profits of Power: Land Rights and Agricultural Investment in Ghana." *Journal of Political Economy*, 116(6), 981--1022.

Greenwood, Jeremy, Ananth Seshadri, and Mehmet Yorukoglu (2005). "Engines of Liberation." *Review of Economic Studies*, 72(1), 109--133.

Heath, Rachel, and Xu Tan (2020). "Intrahousehold Bargaining, Female Autonomy, and Labor Supply: Theory and Evidence from India." *Journal of the European Economic Association*, 18(4), 1928--1968.

Henrich, Joseph, Robert Boyd, Samuel Bowles, Colin Camerer, Ernst Fehr, Herbert Gintis, Richard McElreath, et al. (2005). "'Economic Man' in Cross-Cultural Perspective: Behavioral Experiments in 15 Small-Scale Societies." *Behavioral and Brain Sciences*, 28(6), 795--815.

Henrich, Joseph, Jean Ensminger, Richard McElreath, Abigail Barr, et al. (2010). "Markets, Religion, Community Size, and the Evolution of Fairness and Punishment." *Science*, 327(5972), 1480--1484.

Lundberg, Shelly, and Robert A. Pollak (1993). "Separate Spheres Bargaining and the Marriage Market." *Journal of Political Economy*, 101(6), 988--1010.

Lundberg, Shelly J., Robert A. Pollak, and Terence J. Wales (1997). "Do Husbands and Wives Pool Their Resources? Evidence from the United Kingdom Child Benefit." *Journal of Human Resources*, 32(3), 463--480.

Malinowski, Bronislaw (1922). *Argonauts of the Western Pacific*. London: Routledge and Kegan Paul.

Manser, Marilyn, and Murray Brown (1980). "Marriage and Household Decision-Making: A Bargaining Analysis." *International Economic Review*, 21(1), 31--44.

Mauss, Marcel (1925). "Essai sur le don: forme et raison de l'échange dans les sociétés archaïques." *L'Année sociologique*, nouvelle série, 1, 30--186. English translation by Ian Cunnison: *The Gift: Forms and Functions of Exchange in Archaic Societies*. London: Cohen and West, 1954.

McElroy, Marjorie B., and Mary Jean Horney (1981). "Nash-Bargained Household Decisions: Toward a Generalization of the Theory of Demand." *International Economic Review*, 22(2), 333--349.

Miranda, Veerle (2011). "Cooking, Caring and Volunteering: Unpaid Work Around the World." *OECD Social, Employment and Migration Working Papers*, No. 116. Paris: OECD Publishing.

Ostrom, Elinor (1990). *Governing the Commons: The Evolution of Institutions for Collective Action*. Cambridge: Cambridge University Press.

Oxfam (2020). *Time to Care: Unpaid and Underpaid Care Work and the Global Inequality Crisis*. Oxford: Oxfam International.

Polanyi, Karl (1944). *The Great Transformation*. New York: Farrar and Rinehart.

Rabin, Matthew (1993). "Incorporating Fairness into Game Theory and Economics." *American Economic Review*, 83(5), 1281--1302.

Scott, James C. (1976). *The Moral Economy of the Peasant: Rebellion and Subsistence in Southeast Asia*. New Haven: Yale University Press.

Sen, Amartya (1990). "Gender and Cooperative Conflicts." In Irene Tinker (ed.), *Persistent Inequalities: Women and World Development*, 123--149. New York: Oxford University Press.

Stevenson, Betsey, and Justin Wolfers (2006). "Bargaining in the Shadow of the Law: Divorce Laws and Family Distress." *Quarterly Journal of Economics*, 121(1), 267--288.

Thomas, Duncan (1990). "Intra-Household Resource Allocation: An Inferential Approach." *Journal of Human Resources*, 25(4), 635--664.

Thompson, E. P. (1971). "The Moral Economy of the English Crowd in the Eighteenth Century." *Past and Present*, 50, 76--136.

Titmuss, Richard M. (1970). *The Gift Relationship: From Human Blood to Social Policy*. London: George Allen and Unwin.

Townsend, Robert M. (1994). "Risk and Insurance in Village India." *Econometrica*, 62(3), 539--591.

Udry, Christopher (1996). "Gender, Agricultural Production, and the Theory of the Household." *Journal of Political Economy*, 104(5), 1010--1046.