Introduction

International governmental concern about child labor dates back at least to 1919, the year in which the International Labor Organization was founded, and eliminating child labor is the subject of two of the ILO's fundamental or core conventions. ^[1] The ILO's Convention on the Worst Forms of Child Labor (defined as slavery-like conditions and work likely to harm children's health, safety, or morals) requires "immediate and effective" actions by its signatories "as a matter of urgency." ^[2] A separate United Nations convention identifies the right of children not to be economically exploited and to be protected from work that is "likely to be hazardous or to interfere with the child's education, or to be harmful to the child's health or physical, mental, spiritual, moral or social development." ^[3] Codes of conduct tend to prohibit employment of children younger than a specified age - most commonly the age of 15, with exceptions to the age of 14 in developing countries where ILO practices and local law permit. ^[4]  

Despite clear age limits, measurement in the area of child labor faces serious obstacles because, in general, neither child laborers nor their parents are willing to report violations of those limits, for fear of losing the child's wages. Although child labor has been stigmatized enough in recent years that its incidence is declining, at least in factories used directly by major multinational companies, it remains even in such factories, for example in 2001 in the Kukdong International garment factory in Atlixco, Mexico. ^[5] In agriculture it remains an acute problem internationally, ^[6] and in some cases within the United States as well. ^[7]  

Measurability issues 

Measurement of child labor combines perhaps the clearest numerical criterion in the entire labor standards area - the age of the worker - with some of the greatest disincentives to reliable measurement of that criterion. 

Unlike most other areas of workers' rights, the victims of child labor may not themselves want the violations of their rights to be discovered. Underage workers do not want their pay to stop, and neither do their parents. Other, older workers may also see less immediate self-interest in reporting child labor than in reporting abuses where they themselves could be the direct victim, such as wage cheating or sexual harassment. ^[8] Thus, interviewing workers, chosen either randomly or because they look underage, is less likely to uncover violations than it might be in other areas where the workers being interviewed might see themselves as having more to gain from having violations discovered. 

Documentary evidence like birth certificates or other age-bearing identification papers would be definitive if it were available and reliable, but lack of documents and counterfeit documents can be major obstacles to accurate measurement.  

Existing units of measurement focus far more on workplace policies than on actual implementation. Qualitative approaches play a smaller role, and workers as a source of information show up much less frequently, than they do in other subject areas.  

Current strengths  

Existing units of measurement can effectively capture factory policies on child labor, not just for overall age limits ^[9] but also for important subsidiary issues like post-discovery treatment of child workers to avoid family hardship ^[10] and provide for schooling, ^[11] and special rules for juvenile workers old enough to work under controlled conditions. ^[12] They also cover whether remedial efforts are tracked with records that can be audited, including records of compensatory payments made to the family of a child laborer when the child is removed from the workplace. ^[13] And they cover whether workers learn how the factory implements its child labor policies in practice. ^[14]  

Units of measurement for whether age limits are followed in practice tend to focus on age documentation: whether the factory requires it and makes it available for audit, ^[15] and what a review of it shows. ^[16] Beyond relying on documents, there are a few creative measurement approaches, including using specific cross-examination techniques in questioning workers about their age; ^[17] consulting local outsiders to find out local norms for child labor, which can then be used to guide factory-specific investigation; ^[18] cross-checking with local school records; ^[19] and even noticing if "young-looking" workers "flee" when monitors approach. ^[20]  

Current weaknesses

If an age document is false, or is genuine but being held by the wrong person, existing units of measurement offer little way to detect the fraud. Asking managers if they are familiar with common forgeries is a first step ^[21] but is not enough to measure whether forgeries are actually caught. 

And beyond age documents, existing measurement units are weak. Comparing workers' height and weight with national norms for 15- or 16-year-olds ^[22] is the closest to an objective unit of measurement, but very imprecise. Interviewing workers, management, or local outsiders to learn about child labor in the factory is not prescribed in any detail or with any guidance specific to child labor issues, except to ask about protective conditions for juvenile workers (typically defined as workers between 16 and 18 who are old enough to work, but only in relatively low-hazard jobs). ^[23]  

There is some attempt to use demographics to estimate the potential risk of child labor (such as finding out a worker's father's earnings ^[24] and education level, ^[25] or the number of family members ^[26] ), but without enough methodology to be generally useful. 

In general, existing measurement practice in the child labor area does not try to determine whether there are incentives or disincentives for accurate reporting at the individual factory level, even though the general disincentives are strong. One important exception is checking whether the family of a discovered child worker is compensated when the child stops working, so that child workers and their families will face no or only small financial loss if they admit that the child worker is under age. ^[27]  

Best current practices (selected) 

·        Exploring factory policies on prohibiting child labor and responding to it when it is discovered, including financial compensation to the child's family (see Strengths above)

·        Checking that the factory requires age documentation for employees as a condition of employment ^[28] and keeps the documentation on record ^[29]

·        Checking whether workers understand the prevailing policies and are informed about the factory's actual response in actual cases of violation ^[30]

·        Reviewing dismissal records for a period prior to the monitor's arrival, to see if child workers were cleared out in advance ^[31]

·        Checking for actual tracking of responses in cases here child labor is discovered, including compensation, ^[32] support for schooling ^[33] , and record-keeping on discovered cases ^[34]

·        Checking whether the factory brings in and works with local service organizations when addressing child worker issues ^[35]

·        For juvenile workers supposedly going to school, comparing school hours and work hours, ^[36] consulting school records ^[37] and records of schooling support by the employer. ^[38]  

Although no single source document embodies best practices in toto, one source does stand out for its thoroughness in the child labor area: the Social Accountability International guidelines. ^[39] These and the few but pointed units of measurement on follow-through contained in one confidential source ^[40] are a strong starting point for a general model based on current practice. 

Because of the difficulties in measuring workers' ages directly, these two sources emphasize the indirect indicator of transparency, specifically transparency in how violations of child labor standards are handled once they are discovered. If families can have some confidence that they will be made whole when a child stops working, and that the process will be tracked by overseers of factory management, child workers will have less incentive to lie and hide. Management also will have less incentive to hire child workers in the first place, since compensation plus the wages of the replacement worker for the child will be more expensive than hiring a non-child worker in the first place. The deterrent function of applying these units of measurement - in other words, management's awareness that what it does with child workers will be tracked -- is as important as their function in detecting actual violations. 

Possible improvements suggested by analysis

1.      Require managers to keep track of the number and type of forged age documents that they detect, in the hiring process or subsequently, and to keep the forgeries (or samples) for inspection. This will help monitors learn to recognize prevalent types of forgeries themselves.

2.      Consult with local outside groups not only on child labor norms in the area, but also for any indication of current child labor in the particular facility, common forgery practices, etc.

3.      Consult with local schools for same.

4.      Follow up with local schools in cases of discovered child workers to check on whether remediation is successful, both for the child and for the family.

5.      In discovered cases, when the child involved reaches eligible age, follow up to check whether rehire actually occurs if requested.

6.      Check whether hiring agent and/or personnel manager suffers any negative consequences when child labor hires are discovered.

7.      Ask whether there is special supervision for juvenile workers, and interview the designated supervisor(s).