In 1995, when news broke in the United States that the soccer balls used in kids' games were routinely made by children in other countries, many were horrified. It was not the first time such a scandal had hit the spotlight. Reports were frequent in the 1990s of children working in garment factories, shoe shops, and mines producing goods that would ultimately make their way to U.S. markets. Although child labor is not a new problem, increased trade and more widespread information have brought us closer to practices, problems, and institutions that can conflict with our own beliefs. The intensity of the feelings on the issue was highlighted last December during the World Trade Organization's (WTO) ministerial conference in Seattle. Thousands of protesters clamored for tighter rules on labor in international trade. President Clinton chose the setting to sign an international treaty to eliminate the worst forms of child labor and urged other countries to follow his lead. The United States also called on the WTO to make child labor and other labor rights issues part of the agenda. § But the measure faced stringent opposition from developing nations. Countries such as Thailand, Brazil, and India feared that mandating higher labor standards could rob them of their comparative advantage in cheap labor, price them out of the market, and block their prospects for greater growth and economic development. They argued, in effect, that the industrialized-country focus on child labor used in exports would not only lead to the loss of jobs, but also end up hurting the very children it was intended to help.

Such arguments echoed the heated debates heard in this country less than a century ago. At the beginning of the twentieth century, pressure for federal legislation covering child labor was growing nationally, but especially in the North. It was greeted with resentment in many segments of southern society. They saw it as interference from a richer North which - after having benefited from child labor in its own industrial development - was trying to limit the South's development .

In spite of the controversy, however, the movement to limit child labor prevailed. Today, the number of American children employed full-time has been vastly reduced. But these changes took almost a century and required a host of changes in family income, education policy, production technologies, and cultural norms.

Although the recent scandals have linked child labor to trade, child labor goes well beyond export industries, is widespread, and deeply rooted. Looking at the U.S. experience in reducing child labor can help us get a better understanding of the profound transformation that the elimination of child labor involves.

THE BALL IS IN WHOSE COURT?

The public outcry in the mid 1990s over child labor used to stitch soccer balls put enormous pressure on soccer ball importers such as Reebok, Nike, and Baden Sports. The changes they and other importers made resulted in one of the more successful efforts to eliminate child labor from a specific industry. Consumers can now buy imported soccer balls without feeling as if they are encouraging or benefiting from child labor. But the effect on the children involved is less clear.

The impact was felt most directly in and around the city of Sialkot, Pakistan, a region known for over 80 years for its soccer ball industry. In 1996, 75 percent of the world's soccer balls were made there. The industry was a major source of employment, giving jobs to about 10,000 urban workers and 30,000 rural workers in the surrounding villages. Among them, more than 7,000 children, the vast majority of them between the ages of 10 and 14 (but some as young as five), stitched balls on a full-time basis, according to the U.S. Department of Labor. Other children worked part-time outside of school hours. Middlemen and subcontractors took soccer ball kits to village workshops or households, where workers hand-stitched individual pieces together, glued the bladder to the material, and sewed in the final piece. Once completed, the balls were taken back by the middlemen to the factories for packing and shipment. Children were paid about $0.50 to $0.55 per ball; and depending upon skill and experience, they could be stitching between one and five balls per day.

As a result of the international uproar and bad publicity, Reebok and Nike both contracted with Pakistani manufacturers to make balls in new facilities where all the production was centralized. External monitoring was set up so that no children entered the factory and no soccer ball kits were taken out. The German firm Baden Sports Inc. moved its manufacturing operations to China.

Although well intentioned, these changes were not wholly beneficial to the former child workers and their families. "While Baden Sports can quite credibly claim that their soccer balls are not sewn by children, the relocation of their production facility undoubtedly did nothing for their former child workers and their families,” write economists Drusilla Brown, Alan Deardorff, and Robert Stern. Although Nike, Reebok, and the other firms who switched to centralized stitching centers did not displace work from the region, they reduced the income opportunities for women who formerly stitched soccer balls from home in between other household tasks.

Table: "Working Children Around The Globe"

Brown notes that although the children worked long hours (between eight and nine hours a day), the working conditions were better than in other industries. Most of the children could read, and were not exposed to toxic chemicals or hazardous tools. Moreover, there was no evidence of the practice of intergenerational debt bondage - one of the most blatant forms of exploitation, in which children are given to their parents' "creditor” so that they repay the family's debt through work. In a region that specializes in handmade soccer balls, says Brown, "perhaps learning a trade was the most appropriate use of their time.”

WHAT IS THE ALTERNATIVE?

As the soccer ball example illustrates, if the goal is to improve the welfare of the children, removing them from work is only part of the issue. The larger question is how to provide them with better alternatives.

"Stopping child labor without doing anything else could leave children worse off,” says former Secretary of Labor Robert Reich. "If they are working out of necessity, as most are, stopping them could force them into prostitution or other employment with greater personal dangers. The most important thing is that they be in school and receive the education to help them leave poverty,” he adds.

That children and their families need alternatives was clearly demonstrated in Bangladesh in 1993. Garment employers, fearing the passage of a law in the United States to ban imports made with child labor, summarily fired an estimated 50,000 children from their factories. The Child Labor Deterrence Act did not make it into the law books, but the threat of a boycott was serious enough that garment manufacturers began firing children. Some of the displaced children tried to replace the lost income by engaging in stone crushing, street hustling, and prostitution. In response, international agencies and nongovernmental organizations urged the industry to stop firing underage workers until a safety net was in place. After extended negotiations, the Bangladesh Garment Manufacturers and Exporters Association, the International Labor Organization (ILO), and UNICEF agreed to jointly sponsor schools. They also provided monthly stipends to children to help their families replace lost income. By 1997, over three hundred schools were established that served a total of 9,710 children.

This incident set an example that is now being followed in other interventions. In the case of the soccer ball industry of Sialkot, attempts are being made to provide the displaced children with some form of social protection. An agreement modeled after the one drafted in Bangladesh was reached between the Sialkot Chamber of Commerce and Industry, UNICEF, and the ILO to provide children and their families with informal education for the children and help for the adults in finding other opportunities to earn income. As of August 1999, 5,795 children were attending classes at the 176 centers sponsored.

Still, replacing work with education can be a difficult task. In many developing countries, particularly in remote rural areas, schools are often not available or not affordable for the poorest families. Countries where child labor is prevalent have very limited resources, and, in some cases, governments have not invested sufficiently to have adequate schools in place. In both Pakistan and Bangladesh, special schools needed to be created for the children leaving the soccer and garment industries. In Pakistan, according to the latest figures from UNICEF, only 66 percent of children attended primary school and fewer than half of them reached fifth grade. Bangladesh fared somewhat better with 76 percent of children attending primary school and 61 percent of them reaching fifth grade.

And getting the children to attend schools presents its own challenges, since education is not always seen as something that will be useful. In Sialkot, where the children were offered no stipends, it has required considerable mobilization and awareness raising with the children and their families, particularly if the children had significant earnings.

A GLOBAL PICTURE

Children working in the Pakistani soccer ball industry and in the garment sector of Bangladesh received much attention because they were working in industries that produced goods destined for export to the United States. But children in export manufacturing industries represent only a tiny share of all child labor.

About 120 million children 5 to 14 years of age work full-time in the developing world. If part-time work is included, the number of children working grows to 250 million. Even this figure is probably an underestimate because it excludes unpaid work that does not make its way into the market, such as the work of children - particularly girls - who stay home to do household chores or watch over younger siblings. Probably less than 5 percent of all child workers are employed in export industries in manufacturing and mining, according to the U.S. Department of Labor. The vast majority of children who work - over 70 percent - do so in agriculture.

The best predictor of the incidence of child labor is poverty. Child labor declines steeply as one moves from low-income to high-income countries, notes Princeton economist Alan Krueger. Asia, the most densely populated region of the world, has the largest number of child workers, but it is in poverty-stricken Africa where the highest proportion of children work. About 20 percent of African children aged 5 to 14 work full-time - a little under 40 million children. Eighty million children work if part-time work is included. In countries where 1990 income per capita exceeded $5,000, the employment of children was negligible, says Krueger.

The relationship between poverty and child labor is also true within countries. A range of studies in settings as varied as contemporary Côte d'Ivoire and nineteenth century Philadelphia have found that the incidence of child labor decreases as family income rises. Because of this, many believe that economic growth - if evenly distributed - is a key factor in reducing child labor.

Yet, the issue is not just poverty. Cultural factors also play an important role. In many instances, whether a child works or not depends on such factors as gender, religion, or social caste. In his work on child labor in India, MIT political scientist Myron Weiner pointed out that India's incidence of child labor was higher than that of some countries with lower income per capita. He speculated that religious beliefs and India's hierarchical caste system have prevented education from playing an equalizing role in society. "Those who control the education system are remarkably indifferent to the low enrollment and high dropout rate among the lowest social classes,” he writes. "The result is one of the highest child labor rates in the world.”

In terms of work hazards, children are more susceptible to injuries or work-related illnesses than adults doing the same type of work. The greatest number of illnesses and injuries occur in agriculture, not surprisingly, given the number of children working in that sector. But the likelihood of injury is by far the greatest for children who work in construction and mining, sectors that employ about 3 percent of working children.

Children also tend to work very long hours, leaving little time for school even when it is available. In surveys of 20 countries, the ILO found that in some countries more than half of the children who worked were doing so for nine or more hours per day and up to four-fifths of them did it seven days a week. But evidence on the relationship between school and work is mixed. In a study of child labor in C™te d'Ivoire, World Bank economist Christiaan Grootaeert found that working actually allowed many children to attend (afford) school.

LEARNING FROM THE UNITED STATES

The plight of working children in the developing world today is not very different, and in some cases even less harsh, than that prevalent in countries such as the United States and England during the nineteenth and early twentieth centuries. As recently as 1900, the U.S. Census estimated that 1.75 million children between the ages of 10 and 15 were employed - about 18 percent of the population that age. The majority of these children worked in agriculture. But young children also worked long hours in factories and textile mills, in the anthracite coal mines of Pennsylvania, and in many other industries.

Yet, the contrast between then and now is dramatic. Today, we take for granted that full-time work for children is bad; but in the early eighteenth century, work was believed to be beneficial for a child's character and moral upbringing. The work of children was integral to the agricultural and handicraft economy. Children and not wives were the most common source of family income aside from what fathers earned. They not only worked on the family farm, but also were often hired out to other farmers.

By the early 1800s, with the beginning of the Industrial Revolution, children went into the factory. New machines raised their productivity and also generated a growing demand for their labor. Economists Claudia Goldin and Kenneth Sokoloff estimate that boys and girls under 15 years of age accounted for 23 percent of all workers in manufacturing in the Northeast in 1820. The work of minors was so important to rising industry that Goldin and Sokoloff have speculated that the availability of relatively cheaper labor from women and children gave the Northeast a comparative advantage over the South, and was one of the factors that facilitated for the region's greater industrial development. (Women in the preindustrial South earned considerably higher wages than their counterparts in the North, according to Goldin and Sokoloff, because they were more productive in southern crops, such as cotton and tobacco, than in the hay, grains, and dairy products produced in the North.)

But as children started working in the mills, the nature of their work changed dramatically. No longer were children working on family farms or in small workshops, but rather on factory floors in repetitive motions, for extremely long hours, with dangerous machinery and often abusive supervisors. Public acceptance of children in work began to change. At the same time, the social view of children was shifting and people began to believe that child's play and leisure were not vices but important aspects to healthy development. Also, as infant mortality rates diminished, parents moved from having many children to having fewer, but investing more in their success.

By the end of the nineteenth century, child labor was clearly on the decline. Social opposition to child labor became organized and very active in demanding legislation to limit the employment of minors. The National Child Labor Committee, founded in 1904, pioneered techniques of mass political action. It hired photographer Lewis Hine to capture the poor conditions of children at work and widely distributed his influential photographs.

Major legislative change took place between 1880 and 1910, during which time 36 states set legal limits on the minimum age of workers in manufacturing - 14 years, on average. Finally, a federal law setting a national minimum standard of employment for children was passed in 1938 (see sidebar).

The relative contribution of laws, economic development, and cultural change in reducing child labor is a matter of debate. Between 1880 and 1900, there was a perception that the proportion of children in manufacturing was increasing and this helped fuel the movement for reform. But even though child labor might have increased in the cotton mills of the South and in industries that employed the growing flow of impoverished European immigrants, economists today believe that the fraction of the industrial labor force composed of youth had actually been declining throughout the latter part of the nineteenth century. They attribute this in part to rising incomes, which had made it possible for families to depend less on the income earned by their children. Additionally, new technologies requiring more highly skilled workers led to a decrease in the demand for the unskilled labor of children and to an increase in the returns to education.

Thus, Ohio State University economist Carolyn Moehling believes today that minimum age limits had relatively little effect on the employment of children at the turn of the century. In this view, increasing wealth and changing technology were more directly responsible for the reduction in child labor. Laws helped to cap a process that was already under way. Indeed, the passage of laws banning child labor was possible because there no longer was fervent opposition from parents and industry groups.

But, regardless of whether laws or economic progress led the change, there is widespread agreement that legislation was more effective in bringing about a decline in child labor when combined with compulsory education. Vanderbilt economists Robert Margo and T. Aldrich Finnegan found that, in turn-of-the-century America, compulsory education laws had a significant impact in increasing school attendance only in states that also had child labor laws for children of the same age. They speculate that child labor laws reduced the incentive to seek paid employment and in many cases required children to satisfy educational requirements before starting to work. At the same time, it is easier to monitor and enforce school attendance than to oversee the absence of children from each individual workplace.

The importance of each factor contributing to the decline - laws, rising incomes, and changing production technology - likely varied from region to region and from industry to industry. For example, economists Martin Brown, Jens Christiansen, and Peter Philips found that economic factors played a more significant role than legislation in eliminating child labor in the canning industry in urban areas, but laws had a greater impact in rural ones.

As canneries became more mechanized, they required more steady, reliable labor to maximize the use of the equipment, and children became less desirable as workers. Because of this, child labor declined substantially in urban canneries, which made high investments in capital equipment and whose profitability depended on running the machines continuously and working on a variety of crops year-round. In contrast, rural canneries were smaller-scale and specialized in canning the local crop. They had access to a more limited labor pool and had a much more seasonal schedule of production. Both factors made them highly dependent on child labor during peak production times. Thus, in rural areas more exemptions and qualifications to the laws were passed. But ultimately, the legal restrictions weighed more heavily in reducing child labor there than in urban canneries, where the economics of more capital-intensive production had already diminished its use.

ONE SIZE DOES NOT FIT ALL

Just as the importance of different forces varied across individual industries and regions of the United States, there is a consensus in the field that no one solution to the child labor problem can be applied to all countries. What can be done to eliminate child labor will vary, depending on the level and distribution of income, the availability of education, and the cultural factors that influence child labor.

For the poorest of countries, it may be impossible to really eliminate child labor without income growth or some form of international aid. But for countries with relatively more resources, a range of policy alternatives can help bring about change. "Government can intervene in the market to create a variety of incentives, such as providing better and more schools, giving school meals, and improving conditions in the adult labor market, which result in a reduction of child labor,” says Cornell economist Kaushik Basu.

What is to be done depends on the educational infrastructure in place. The American example clearly showed that education played a key role. Better educated workers in the United States were more productive and thus were better able to provide for their own children, reducing the need for child labor. And, taking children out of the labor force may help them only if you provide them with a better alternative.

Export goods are only the tip of the iceberg when it comes to child labor. Perhaps international pressure will help bring about a change in the perception of child labor in countries where it is accepted as an integral part of life. But, for the permanent elimination of child labor, a cultural change ultimately has to come from within developing countries.

Looking to the past gives hope for the future. Child labor, once endemic in countries like the United States and Great Britain, has been greatly reduced there. The process was long, and involved sustained change on many different fronts. But the payoff was worth the effort, for it is hard to build a strong future with such small hands.

DOMESTIC RESTRICTIONS ON CHILD LABOR

The first law in the country to place limitations on the employment of children was enacted by Massachusetts in 1836. The driving concern was not the age of working children, the hours they worked, or the danger involved in their occupations. Rather, state legislators were worried about work's impact on children's education. The Massachusetts law stipulated that no children under age 15 could be employed unless they had attended school for at least three months in the preceding year. Education was later made compulsory in Massachusetts in 1852. Compulsory schooling laws were enacted in most of the states outside the South during the last half of the nineteenth century, according to William M. Landes and Lewis C. Solomon.

Massachusetts was also first to limit children's workdays, banning children under 12 from working more than 10 hours in 1842. Pennsylvania established the first minimum age in 1848 12 years for work in cotton, woolen, and silk mills. The first night-work provision was enacted in 1888, when Massachusetts prohibited children under 14 from working between 7 p.m. and 6 a.m. A few, mostly Northeastern states, followed these legal examples, setting minimum age and other requirements according to their local needs and preferences.

Enforcement of these laws was scant. No proof of age was required for employment, and it wasn't until 1867 that Massachusetts instituted the nation's first factory inspection. Even so, the wording of the laws gave ample space for loopholes. For instance, "in Massachusetts, New Jersey, and Rhode Island, the only punishable violations of child labor were those committed ‘knowingly,'” according to historian Walter Trattner.

In 1879, only seven states restricted the age of children in manufacturing, with an average minimum age of 11. The uneven formulation of state laws led reformers to push for a national standard. But this turned out to be a drawn-out fight. Two federal laws, the Keating-Owen Bill of 1916 and the Pomerene's Tax Bill of 1919, were declared unconstitutional in 1918 and 1922, respectively, by the U.S. Supreme Court, which considered them in violation of states' rights. Beginning in 1922, anti-child-labor reformers pushed for a constitutional amendment intended to grant the Congress specific authority to legislate protections for working children.

Most of the opposing votes in Congress came from southern textile states, and it was the Southern Cotton Manufacturers, an industry group, which contested the constitutionality of the laws. Southern states had a larger agricultural base and, particularly after the Civil War, were afflicted with widespread poverty - both factors associated with a greater incidence of child labor. Also, the textile industry, which relied heavily on children, spread and grew dramatically in the South during the second half of the nineteenth century, particularly in Georgia, Alabama, and the Carolinas.

With the Great Depression, the national mood on federal child labor legislation changed. It seemed to many that adult workers were being replaced by young children at lower wages, and this helped shore up the support for restricting child labor.

In 1938, with the passage of the Fair Labor Standards Act, a national standard on the employment of children was finally set. The FLSA prohibited interstate commerce in goods made by children under 16 years of age and by children under 18 years in particularly hazardous occupations. Exceptions could be made for 14- and 15-year-olds in occupations other than manufacturing and mining, if the work did not interfere with their schooling, health, or well-being. Still, the law did not cover children employed in agriculture during their school vacations or migratory child workers.

PHOTOGRAPHY FOR CHANGE

Lewis Hine's camera was one of the most powerful weapons mustered by the child labor reformers in the early twentieth century. Employed full-time by the National Child Labor Committee (NCLC) between 1908 and 1918, Hine investigated and photographed the working conditions of children in industries in most states east of the Mississippi River and in Missouri, Colorado, Texas, and California. He was among the first to use photography to promote social change, and his images left a lasting impression.

The match between the NCLC and Hine was rather fortuitous. Hine was born in Oshkosh, Wisconsin, in 1874. He started out his professional career as an educator and began using photography as an educational tool, through his acquaintance with Frank A. Manny, a professor of education and psychology, at the Ethical Culture School in New York City. Both men were interested in studying the immigrant tide flowing through Ellis Island at the turn of the century, and Manny assisted Hine in taking some of his first successful pictures.

In the meantime, Edgar Gardner Murphy, an episcopal clergyman in Montgomery, Alabama, and the founder of the Alabama Child Labor Committee, had pioneered the use of photography in the anti-child-labor campaign. Around 1901, Murphy had smuggled a camera into some of Alabama's textile mills and produced a powerful pamphlet called Pictures from the Mills. When the child labor committees in Alabama and New York joined forces and founded the NCLC in 1904, Murphy was one of the movement's most influential figures. In 1906, Hine began photographing children at work for the NCLC on a freelance basis, and by 1908 his passion for photography replaced his teaching career.

The images Hine captured for the NCLC left a deep impression not just because of the emotional topic, but also thanks to Hine's aesthetic sensibilities. In order to obtain the photographs, Hine defied difficult circumstances. His photographic equipment - a modified box-type 5" x 7" camera - weighed up to 50 pounds and the explosive magnesium flashpowder used in interior settings could be quite dangerous. (Apparently, Jacob Riis, a famous contemporary social reform photographer, twice set the places he was photographing on fire.)

Hine's entry was not welcomed at the mills, mines, or factories. He often had to pass as a fire inspector, an insurance salesman, a Bible salesman, or an industrial photographer. To document the specific cases of child labor, he surreptitiously wrote notes in a notebook he kept in his pocket, and he used the buttons on his jacket to measure the children's height. He directed his attention to the particular industries or trades that the NCLC targeted coal mines, glass factories, Southern and Northern textile mills, seafood canneries, street trades (newsboys and messengers), and agriculture. All in all, he took over 5,000 photographs for the NCLC.

By 1918, the NCLC decided to focus more on the investigative work than on photography, and Hine moved on to other projects. He photographed the Red Cross's efforts at the end of World War I, created a series on men at work, participated in photography projects during the New Deal, and documented the construction of the Empire State Building. But commissions for his work declined in the last decade of his life, and he died in poverty in 1940.