A Follow-up Study of Child Hunger in Canada - June 2001

5. Discussion

Previous

Contents

Next

"The era of globalization is upon us " (Anderson and Cavanagh, 2000). Globalization has supplanted the old ways of Fordist capitalism and its post-WWII emphasis on Amass production for mass consumption" (Cox, 1987). In this new era of intensive global economic restructuring, corporations have trimmed their workforces, sub-contracted substantial proportions of their production, and re-settled in the more than 200 export processing zones around the globe (Bernard, 1994).

The era of globalization has demanded a new public policy framework throughout the world. At its core, this policy framework for globalization — typically called the neo-liberal policy agenda or the Washington consensus — emphasizes the principle of free markets, including liberal trading regimes, and undistorted labour markets. In accordance with the principle of free markets, states have privatized public operations, removed barriers to trade, lowered corporate taxes and other investment hindrances, and, most importantly, reduced income supports (such as social programs and wage legislation) that are seen by some to distort labour markets (Greider, 1997).

In keeping with the demands of globalization, the neo-liberal public policy framework is taking shape in Canada. The Canadian state is no longer content with the Keynesian policy framework — counter-cyclical spending, significant state involvement in the economy, and generous social transfers — that served governments in the past. The centerpiece of the neo-liberal agenda in Canada is the North American Free Trade Agreement. Successive federal and provincial governments have also cut back on social program spending, especially welfare services and Employment Insurance, reduced corporate taxes, implemented the Goods and Services Tax (GST) and Harmonized Sales Tax (HST), aggressively promoted entrepreneurialism, attacked unions and labour legislation, privatised state-owned enterprises and, lastly, re-structured and downsized government (Laxer, 1998). These policies have had the cumulative effect of driving down the social wage in Canada, thereby lowering production costs for transnational corporations (Campbell, Gutierrez Haces, Jackson, Larudee and Sanger, 1999).

The corporate hubris regarding globalization sits awkwardly against the culture of austerity felt by so many Canadians. Despite optimism about global growth having a net positive impact on Canada, we have seen the expansion of poverty throughout the country, falling real wages and a growing gap in incomes between the wealthiest and the poorest segments of Canadians (Yalnizyan, 2000).

Within this generic policy context then, let us examine the results of this study and its implications for Canadian social and health policy.

5.1 Child Hunger in Canada

According to the families of the NLSCY, the extent of the problem of children experiencing hunger is little changed between 1994 and 1996. While both the percentage and number of families reporting hunger increased, the increases failed to reach statistical significance. The prevalence rate of 1.6% of NLSCY families reporting hunger is about 25% of population-adjusted rates found in the United States, which is the only available comparator. The Third National Health and Nutrition Examination Survey (NHANES III) in the United States, conducted between 1988 and 1994, attempted to measure food insufficiency for the first time among its nationally representative sample population (Alaimo, Briefel, Frongillo and Olson, 1998). The question asked of respondents was whether or not they "sometimes" or "often" did not get enough to eat. Results revealed that 6.8% of families with children 2 months to 5 years, and 5.7% of families with children 6-11 years, lived in families that reported sometimes or often not getting enough food to eat. In 2.7% of the families surveyed, children younger than age 17 had cut the size or skipped meals in the previous month because of a lack of food.

The Community Childhood Hunger Identification Project (CCHIP) categorized families with children as hungry based on positive parental responses to five of eight standardized questions on the experience of food insufficiency. Data from nine U.S. states revealed that 8% of children under the age of 12 experienced hunger (Wehler, Scott, and Anderson, 1996.) There is only a sparse literature on food insecurity in the United Kingdom, New Zealand, or Australia (Dobson, Beardsworth, Keil, and Walker, 1994; Dowler and Calvert, 1995; Wilson, 1997; Lang, 1999; Uttley, 1997), and very little is published on this topic in the European Union. So while the Canadian rate is better than that seen in the United States, it is plausible that the Canadian rate is still well above European rates given that Canadian poverty lines are typically much closer to U.S. levels than continental levels (United Nations Development Program, 1999).

Public perception of child hunger in Canada is actually much greater than the NLSCY statistics show. The "National Child Hunger Survey" (Thompson Lightstone and Co. Ltd, 1997), commissioned by the Canadian Living Foundation and conducted by a consulting firm in 1997 surveyed 2000 randomly chosen adults. It found that respondents believed that 42.2% of Canadian schoolchildren were not eating an adequate breakfast. When asked to rank child hunger in terms of importance compared with other national issues such as quality education, national unity, quality health care, and unemployment/job creation, between 85% and 89% of respondents stated that they believed that child hunger was at least as important as these other issues.

Canadian's uncontested belief that child poverty equals child hunger, as demonstrated by the National Hunger Survey, encourages an exaggerated view of the occurrence of child hunger in Canada. In the NLSCY, 24.6% of children aged 0 to 11 lived in poverty, defined as families living below Statistics Canada's low income cut-off (Ross, Scott and Kelly, 1996). In 1996, less than one- fifteenth of these families reported that their child experienced hunger because there was no food in the house or money to buy food. Although the incidence of hunger in children is disturbing and in need of public solutions, this study and others (Rose, 1999) demonstrate that it is not necessarily a consequence of poverty.

Rose emphasizes the importance of using direct measures of food security because indirect measures of well-being such as poverty-level incomes are neither specific nor sensitive to the hunger condition (Rose, 1999). The NLSCY uses a direct parental report of hunger. In terms of the actual question used in the NLSCY cycles, the question "Has your child EVER gone hungry" was interpreted as changeable and in fact was sensitive to at least a biennial change in hunger status.

Is a self-report of hunger valid? The answer appears to be "yes". Rose and Oliveria (1997) reported that in the United States, at least for adults, self-reported hunger measures are valid surrogate measures for low intakes of required nutrients. Their study adds further credibility to national surveys using self-report measures of hunger in the United States such as the Community Childhood Hunger Identification Project and the Food Security/Hunger Module of the 1995 Current Population Survey, whose questions are similar to the NLSCY (Sidel, 1997).

5.2 Who Becomes Hungry?

Campbell's conceptualization of food insecurity states that food insecurity results from anything that "limits either the household resources (money, time, information, health, etc.) or the proportion of those resources available for food acquisition. Hence risk factors include factors that limit employment opportunities, wage and benefit scales and social assistance benefits, or that increase non-discretionary non-food expenditures such as the cost of housing and utilities, health care, taxes, child care and the likelihood of emergencies" (Campbell, 1991). Among the families of the NLSCY, the independent predictors of hunger were lone parent-led and aboriginal households, low household income, higher number of siblings, and PMK reports poor or fair health. These findings are not surprising given the 61% poverty rate seen in lone-mother led households, and the fact that overall, 52% of aboriginal households are classified as poor (CCSD, 1999).

Frequent hunger was predicted by low household income, a higher number of siblings, and low maternal education. In Tarsuk and Beaton's (1999a) study of food bank recipients, the only socio-demographic variable associated with severity of food insecurity was higher number of children in the household. The observation of higher number of siblings predicting both frequent and any hunger may be related to the burden of literally 'another mouth to feed'. The main income feature of frequently hungry households was a very low annual mean personal income of the PMK which was in fact significantly lower than the already low annual mean personal income of the PMK in occasionally hungry families.

Income is clearly one of the most important determinants of food insecurity and hunger. For example, the 1995 U.S. Current Population Survey (CPS) showed that 17% of households with incomes less than 50% of the poverty level were hungry, whereas only 1.4% of those 185% or greater above the poverty line were so affected. Rising incomes have also been linked directly to declines in food insufficiency between 1988 and 1994 (Rose, 1999).

The clear connection between income and food insecurity and hunger raises concern about falling real wages in Canadian society, and especially about the stagnation and decline in the real minimum wage rate in all of the provinces. Between 1976 and 1995, for example, the minimum wage rate fell by more than 25% in eight of the country's ten provinces (Schellenberg and Ross, 1997, p 42). In our study, 63% of hungry households received employment income over the year, and this was the main source of income for 54% of hungry households. The universal decline in the minimum wage rate is one of the key contributing factors to the expansion of the working poor. The growth of the working poor, quite simply, is likely to exacerbate food insecurity and hunger.

The situation was equally bleak for recipients of social assistance or welfare; 41.9% of families who reported hunger in 1996 received social assistance or welfare as their main source of income; and among the frequently hungry, social assistance was the main source of income in 61.0% of the cases (data not shown). The National Anti-Poverty Organization report entitled, "Poverty and the Canadian Welfare State: A report card" (NAPO, 1998), outlines the erosion of the Canadian welfare state between 1990 and 1996 including the tightening of eligibility requirements for social assistance and/or cuts in benefits. These years saw the highest level of income inequality in Canada in twenty years with the poorest 20% seeing their average incomes fall by $500 as a result of decreased government transfer payments and lower real labour market earnings. The creation of the Canada Health and Social Transfer in 1996, a move that reduced federal government transfers to the provinces and territories for health, education and social services, has likely aggravated these effects by making "it all but impossible for provinces and territories to make long-overdue improvement in the welfare systems." (National Council of Welfare, 1999-2000, p. 67). The conclusion as of 1996 was that real suffering and increased hardship had occurred for low income Canadians as a result of continued erosion of social assistance rates.

Welfare rates do recognize the presence of children within households, although the overall rates are insufficient and fall well below the poverty line.In contrast, wages in Canada do not recognize family size and the number of children. The insensitivity of wages to family type — put bluntly as the number of mouths to feed — contributes to low income levels in many families and by implication, as this study confirms, is bound to contribute to food insufficiency.

With the second cycle of the NLSCY, we were able to track hungry families over time as experiencing hunger in 1994 but not in 1996 (moving out of the hunger state), experiencing hunger in 1996 but not in 1994 (moving into the hunger state), or experiencing hunger in both 1994 and 1996 (persistent hunger state). Only 22.4% of families reported persistent hunger.

Persistently hungry families had the least change in socio-demographic variables, i.e., the most stable educational levels, family structure, and employment/unemployment status, of all three groups. Families that moved out of the hunger state reported significant increases in income adequacy, and a mother who gained full time work compared to the other two groups.

The risk of moving into the hunger state was 5.75 times higher for families when at least one additional sibling was added to the household, and 5.64 times higher for families where the father lost full time work, than for families that moved out of the hunger state or who remained persistently hungry. Other risks for moving into the hunger state included adding another parent to the household, father's improved educational status (perhaps a result of returning to school after job loss), and mother's unemployment status change. A precipitous decline in mean annual household income (-$2690) also occurred in families that moved into the hunger state.

The fluidity of hungry families is consistent with recent policy trends in Canadian society that have created dislocations in work, and have increased restrictions on access to social assistance and Employment Insurance, making it harder for families to "get by" at times of difficulty or when the family is stressed with a new addition to the household. Our finding regarding the relationship between hunger and precipitous income decline is alarming given the direction of public policy and economic trends over the last two decades, directions likely to encourage sharp losses in income. It is significant that seasonal employment and the tendency towards a casualized labour force — part-time and temporary employment — continue throughout Canada, especially in the Maritimes. Studies have shown that seasonal and temporary workers are susceptible to sharp declines in income over the course of the year (L'Italien, LeBreton and Grignon, 1999). More generally, the decline of the "good job" characterized by long-term, stable employment, and the highly touted transition to the so-called "flexible worker" is likely to encourage a sudden loss in income over the course of adult life (Rifkin, 1995).

5.3 Hunger and Health

Both in 1994 and in 1996, health status of the both the PMK and the child were significantly related to hunger as were activity limitations, and the presence of a chronic condition. Migraine headache, a condition that can be exacerbated by stress, and chronic back problems, were significantly higher in hungry compared with other PMKs. A deterioration of the health of the PMK was 3.5 times more likely among those who moved into the hunger state than among those who either persisted in, or moved out of, the hunger state.

The causal relationship between food insufficiency and ill-health of parents in hungry households is unclear. Cristofar and Basiotis (1992) found that lower levels of intake were associated with reported food insufficiency among women and that food insufficiency of women in relation to other members of the household was related to larger household size, mothers who reported poor health, and women who were smokers.

Hunger does seem to impact child health. Child health status was found to be negatively affected by hunger, an effect that was mediated by gender. For boys, poorer health status occurred at the time of hunger; for girls, the longitudinal experience of hunger over two time periods, had a significant negative impact on their health. The literature documents that nutritionally disadvantaged children have more health problems such as anemia, weight loss, frequent colds, and infections than nutritionally advantaged children (Maxwell and Simkins, 1985; Shah, Kahan, and Krauser, 1987; Miller and Korenman, 1994). Overall nutritional intake is also inadequate and specific deficiencies are more likely in the face of inadequate dietary intake (Rose and Oliveira, 1997). Insufficient food is associated with impaired growth and mental development (Brown and Pollitt, 1996).

Asthma is becoming recognized as a condition related to low socioeconomic status (Wissow, Gittelsohn, Szklo, Starfield, and Mussman, 1988; Weitzman, Fortmaker, Walker, and Sobol, 1989;). Children of hungry families were affected not only by higher rates of asthma but higher severity, or poorer control of asthma, as indicated by disproportionately higher reported Ventolin use among hungry asthmatic children compared with non-hungry asthmatic children.

Cigarette use in the households of hungry families is partially responsible for higher child asthma rates (Chen, Rennie and Dosman, 1996), but also represents nicotine addiction among their mothers who are likely stressed, and possibly hungry (Stewart, Brosky, Gillis, Jackson, Johnston, Kirkland, et al, 1996). Smoking produces harmful effects to health and extra cost in financially stressed households (Mummery and Hagen, 1996). The 1996/1997 National Population Health Survey reported on the smoking behaviour of Canadians and found that 26% of women over the age of 15 were smokers (NPHS Highlights, 1999). At a smoking rate of 58%, hungry mothers are clearly smoking well above societal norms.

Behaviour and school performance could not be assessed for children reporting hunger because there were no questions on these domains asked of the entire cohort. Nutritionally disadvantaged children have been shown to have higher school absences, and impaired learning compared with other children (Wehler, Scott, and Anderson, 1992; Skolnick, 1995). Psychosocial functioning, manifested by behavioural and attention problems in school, has been shown to be impaired in children of low income families experiencing hunger (Murphy, Wehler, Fagan, et al., 1998). The Community Childhood Hunger Identification Project studies revealed that children who were classified as hungry were more likely to have mood and attention problems and more likely to be absent from school than poor children who were classified as not hungry (Wehler et al, 1996). Healthy growth and development in general are impaired in the face of food insufficiency, and children are less likely to succeed as healthy citizens over the long term as a result (Wachs, 1995). In the future, it may be possible to derive a learning, behaviour, or vulnerability index from age-group scores using such variables to determine outcomes for hungry children of the NLSCY.

Children's weights and heights were reported by the PMK. From these, the body mass index was derived for children ten years of age and older. The resulting sample size was small so that findings are speculative. The longitudinal results demonstrated that a hungry girl child's weight is heavier in the face of persistent hunger, while for a hungry boy child, it is lower than reported for non-hungry children.

A hunger study of midwestern U.S. inner-city children compared anthropometric indicators for children who were classified as hungry with those classified as at risk for hunger, and those classified as non-hungry (Cutts, Pheley and Geppert, 1998). No differences were observed among these three groups for any standardized growth data including weight for age. The study recognized that the average weight for height of the population was higher than expected, and in line with heavier weights seen in low income populations (Centers for Disease Control and Prevention, 1995). The finding of weight discrepancies between girl and boy children exposed to persistent hunger bears further scrutiny.

5.4 Coping with Hunger and Food-Bank Use

There were few differences in responses to lack of food and coping strategies between 1994 and 1996. As seen in the past, parental deprivation was about six times more frequently reported than child deprivation. This has been well-documented elsewhere (Radimer, Olson, Greene, Campbell, and Habicht, 1992; Cristofar and Basiotis, 1992).

The 1996 results confirmed differences seen in 1994 between those who use food banks as a mechanism of coping with food insufficiency compared with those who seek help from relatives or friends. Food bank use seems to be a genuine marker of food deprivation, a strategy of last resort. While food bank use may not be a specific marker of food insufficiency (about two-thirds of the hungry in this study do not seek food bank support), it seems to be highly sensitive to the hunger state (i.e., few people who use a food bank are not truly hungry) [Tarasuk and Beaton, 1999a].

Seeking help from a food bank was significantly more likely among recipients of social assistance or welfare, lone-parent led families, and among families with a higher total number of siblings in the household. In a study of food bank users, non-users and past users in a low-income population of single mothers, the main differences between food bank users and non-users were larger family size, and severity of hunger (Smith and Hoerr, 1992). Residents of Ontario were significantly most likely to use a food bank in 1996, as they were in 1994, although this variable was not an independent predictor of food bank use in logistic regression analysis.

Food banks were initially established as an emergency response to a perceived hunger crisis. Food banks are now accepted as part of an institutionalized, albeit voluntary, network of food distribution (Campbell, 1991). It is important to emphasize that food banks have not been incorporated into public policy and are not a policy solution. The inadequacy of food banks as a source of quality food assistance has been illustrated by a recent study by Teron and Tarasuk (1999) who examined food bank recipients' hampers in Toronto in June and July 1998: food amounts were small; damaged or outdated goods were common; and food quality was poor.

In view of the limitations of food banks, concern must be raised about the growing trend of governments to "get out" of the business of helping people. Contemporary policy trends seem destined to worsen social programs and encourage the growth of private voluntary efforts to relieve poverty. Our concern is twofold and is not, of course, with the generosity of the innumerable, well-intentioned people who indefatigably offer their services on behalf of those in need, but rather: (a) questions the ability of these private efforts to achieve the necessary standards that could be reached with properly funded, government-run programs, and (b) flags the general problem that charities may unintentionally relieve governments from providing effective social services that eradicate problems such as poverty and food insufficiency in the first place.

5.5 Hunger and Family Dysfunction

With two NLSCY cycles collected, we are now able to determine longitudinal effects of hunger on family function. Higher family dysfunction was found as a direct result of hunger, as well as a longer-term effect of hunger persisting over time. This would imply that the hunger state, either at one time, or persistently, leads to family stress. The result is greatest for families with girl children. Hamelin and colleagues (Hamelin, Habicht and Beaudry, 1999) studied the consequences of food insecurity in 98 low income households in Québec. They identified food insecurity as causing "a variety of sociofamilial perturbations & disrupted household dynamics as well as distorted means of food acquisition and management." These findings support our results related to family dysfunction as both a direct and longer-term consequence of hunger.