Mothers’ Milk and Measures of Economic Output

Abstract

Thoughtful economists have long been aware of the limitations of national accounting and GDP in measuring economic activity and material well-being. Feminist economists criticize the failure to count women's unpaid and reproductive work in measures of economic production. This paper examines the treatment of human milk production in national accounting guidelines. Human milk is an important resource produced by women. Significant maternal and child health costs result from children's premature weaning onto formula or solid food. While human milk production meets the standard national accounting criteria for inclusion in GDP, current practice is to ignore its significant economic value and the substantial private and public health costs of commercial breastmilk substitutes. Economic output measures such as GDP thus are incomplete and biased estimates of national food production and overall economic output, and they distort policy priorities to the disadvantage of women and children.

Keywords:

• Breastfeeding
• national accounts
• health
• food supply and demand analysis
• childcare

Notes

JEL Codes: I120, J160, E100

The term gross domestic product (GDP) used in this paper is interchangeable with gross national product (GNP) for most countries. The main difference is that GNP adjusts for income earned or remitted overseas. National accounting practice is moving to the use of GDP.

Such changes included attention to working conditions affecting women's ability to combine breastfeeding and work, such as maternity leave and benefits, nursing breaks, and childcare facilities.

This research points to “strong evidence that human milk feeding decreases the incidence and/or severity of diarrhea, lower respiratory infection, otitis media, bacteraemia, bacterial meningitis, botulism, urinary tract infection and necrotizing enterocolitis.” The American Academy of Pediatrics (AAP) also cited a number of studies of human milk feeding showing a possible protective effect against sudden infant death syndrome, insulin-dependent diabetes mellitus, Crohn's disease, ulcerative colitis, lymphoma, allergies, and other chronic digestive diseases. Since the AAP study, several important, high-quality epidemiological studies have provided further evidence of the health risks of artificial formula feeding, including long-term risks of obesity, high blood pressure, and heart disease as well as pneumonia, gastroenteritis, respiratory illness, allergies, and necrotizing enterocolitis.

One argument is that the economic value of breastfeeding as a food commodity derives from the nutritional and immunological (biochemical) properties of human milk (Clemens Kunz, M. Rodriguez-Palmero, Berthold Koletzko, and Robert Jensen 1999) and that breastfeeding contributes to the quality of childcare services by enhancing mother – child bonding, security, and attachment (Kennell and Klaus 1998) and advancing jaw and speech development (Miriam Labbock and G. E. Hendershot 1987).

Countries where human milk banks operate are an exception. Here, however, SNA93 would count human milk as the value of production by milk banks, or equivalently what is expended on the milk by households, rather than as a transfer of economic value by households producing and donating a food product to milk banks.

Information on the operation of human milk banks in North America, Europe, and Asia is provided in the following: Mary Rose Tully (1991); Armida Fernandez, Jayshree Mondkar, and Ruchi Nanavati (1993); Lois Arnold (1994, 1996); Skadi Springer (1997); D. Gutierrez and J. A. de Almeida (1998); Human Milk Banking Association of North America (2003); United Kingdom Association for Milk Banking (2004).

For example, to estimate the cost of replacing human milk from recent declines in breastfeeding in Chile, Kenya, Singapore, and the Philippines, Alan Berg (1973) used data on breastfeeding prevalence to estimate national human milk output. He then measured its economic value using a price of US$240 per ton of formula. The study by Ted Greiner, Stina Almroth, and Michael C. Latham (1979) for Ghana and the Ivory Coast estimated the value of national human milk production by calculating the local cost of the formula and bovine milk that would be necessary to provide the equivalent caloric value if breastfeeding mothers switched to artificial feeding. Likewise, Jon Eliot Rohde (1974) used the avoided cost of purchasing cow's milk for Indonesia's 1- and 2-year-olds to calculate that the value of extended breastfeeding equaled 80 percent of the country's health budget.

The market price of artificial formula milk will be an underestimate of its true price unless consumers possess perfect knowledge about the nutritional and health impacts of not breastfeeding and make informed choices; the infants' feeding “preferences” are meaningful, forward looking, and faithfully reflected in decisions taken by their caregivers; and there are no societal “externalities” in the production or consumption of breastmilk. Many health risks of not breastfeeding are borne by the community (and the infant), not the parent. Gaps or time lags in accumulation of knowledge about the adverse health and development consequences of formula feeding mean that the market price will understate the true cost. The mother or caregiver, as “agent” for the baby, may also not properly reflect the infant's preferences or “willingness to pay” in her purchasing decisions. On the other hand, the price of formula may be inflated by the market power of commercial baby food companies. But given that the health costs of formula feeding are very large, it is more likely that the market price of artificial formula understates its economic cost than overstates it.

All exchange conversions in this study are at a rate of A$1 = US$0.75.

The preferred estimate, using the market alternative approach, is based on the value of human milk traded by milk banks in Norway of US$50 per liter (Oshaug and Botten 1994). A similar figure resulted from using alternative valuation approaches (Smith 1999). The economic value of breastfeeding also depends on how the time and commodity cost of breastfeeding is assessed. For example, breastfeeding mothers may consume more food, or formula feeding may reduce the time she needs to spend with her infant. For a full discussion of these issues and estimates based on time-use surveys of infant feeding and lactation energy needs, see Smith, Ingham and Dunstone (1998).

For a discussion of valuation issues, see Smith (1999: 76 – 80), which acknowledges that the price of banked milk reflects the costs of supply and the particular economic and institutional characteristics of a small and restricted market.

Using this method, Smith (1999) could compare estimates of the value of Australia's human milk production with those for Norway for the same year. While the price of expressed breastmilk used in the study has some practical and conceptual problems, it is a more accurate representation of the economic value of human milk than the price of formula: the price that health providers will pay for human milk is likely to reflect a relatively informed view of its health benefits. The market alternative method is also most consistent with national statistical procedures for valuing market production, and results can be directly compared with national accounting aggregates such as GDP (Australian Bureau of Statistics 1992) and with estimates of the economic value of unpaid work in accordance with the international convention for “satellite accounts” of unmarketed household production (Australian Bureau of Statistics 1990).

For replacement cost valuation, the study used the official wage for childcare workers in Australia in 1992 of around A$13 per hour. The cost of employing three such workers for an eight-hour day and estimates for a wet nurse's average daily milk production per shift of 1.875 liters (Wickes 1953) implied an approximate cost per liter of replacing mothers' milk of A$55. For the opportunity cost estimate, the study took the wage rate of A$11.16 per hour to approximate the value of nursing mothers' time; this is the wage rate used by the Australian Bureau of Statistics (1992: 23) for valuing “other housework.” In this case, the shadow price for donated human milk in Australia would be around A$75 per liter. For detailed discussion of these assumptions, see Smith (1999: 79 – 80).

While there are welfare gains from treating illness, the high resource costs of treatment make this an inefficient way of achieving good health compared to prevention through human milk feeding.

While risks of toxins in breastmilk, HIV infection, or drug use do not generally override health recommendations for exclusive breastfeeding, acknowledging women's milk productive capacity as an asset also reinforces the economic harm from environmental damage and public health problems.

For example, an increase in dairy herds to provide bovine milk supplies represents an unnecessary resource cost. In Australia, each dairy cow typically requires 0.77 hectares of land to produce around 5,000 liters of milk annually. Hence replacing the 32 million kilograms estimated annual production of human milk with artificial formula milk powder, equivalent to 238 million liters of milk, requires the use of around 37,000 hectares of prime farming land, which could be used for other productive purposes.

Many mothers combine continued lactation with paid employment, especially those with sufficient maternity leave to allow proper establishment of lactation, those with flexibility in their working arrangements and/or in their timing of return to work, or those with an older baby requiring less frequent milk feedings. However, there is clear evidence that a return to work before six months reduces the average duration of breastfeeding (see Judith Galtry 2002 for a succinct review). Galtry points out that inequality in access to either family or medical leave is a barrier to increased breastfeeding rates in the United States, while welfare reform, which forces lone mothers back into the workforce soon after birth, further reduces breastfeeding among lower income and disadvantaged groups. For mothers who wish or are compelled to return to work soon after childbirth, workplace measures (such as “phasing back,” part-time work, shorter working days, and flexitime) are also necessary to allow integration of work and family commitments.

The conceptual basis for such a calculation is implicitly acknowledged in research by the Australian Treasury on the nation's public investment performance (Peter Depta, Frank Ravalli, and Don Harding 1994), which proposes that increased public investment in human capital through certain health and education expenditures be seen as an offset to the slower expansion of public investment in physical capital in recent years.

Realizing this potential gain, however, would require institutional and policy change to ensure that mothers in disadvantaged socio-economic groups have comparable access to breastfeeding support and flexible employment. Although rural poor mothers in developing countries are more likely to breastfeed than urban middle-class mothers, the reverse is generally true in Australia (Susan Donath and Lisa Amir 2000).

The newly introduced goods and services tax (GST) applies to the hire or sale of lactation aids such as breast pumps, which are used by employed mothers, those with premature babies, or those with breastfeeding problems to maintain the production of milk for their babies, or by some mothers to donate milk for others. That is, expressed human milk is “input taxed.” On the other hand, commercial baby foods and formulas are free of GST at all stages of production and sale. Baby food and formula manufacturers can also claim input tax credits for GST against GST paid on-farm milking machinery or infant food manufacturing equipment.

Additional information

Notes on contributors

Julie P Smith

JEL Codes: I120, J160, E100