This form of underemployment - a literacy surplus problem - is as much a concern as the literacy deficit problem (inadequate levels of literacy among workers) that has captured public attention in the past decade. Both problems should be addressed within the same fit-mismatch framework. At one end of the continuum are workers whose literacy skills fall well below the minimum requirements in most jobs. At the other end are the highly literate employed in jobs that frequently under-use their skills. In the first case, potential human resources are not being developed. In the second, the human capital available is not being optimally used. As a result, some of it may be lost (Krahn, 1997).

Thus, it is essential to focus not only on persons with skill deficits but also on persons whose skills surpass the requirements of the job. For workers with moderate or high levels of literacy, the long-term effect of working under such circumstances could be loss of skills. For workers with low literacy levels, an unchallenging work environment could reduce the likelihood of their developing literacy skills either on or off the job.

To what extent do employed Canadians use their literacy skills at work? Defining optimal use of a society's human capital (in this study indexed by literacy skills) is difficult, because sometimes workers change jobs or jobs change in their skill requirements. All the .same, a better fit would be preferred over a poor fit (see Constructing measures of literacy fit and mismatch in the workplace.) Ideally, public policy coupled with market incentives would induce employers to increase workplace literacy requirements (that is, to create more knowledge-based jobs), and to encourage employees with lower literacy skills to upgrade through further education and training.

Regarding prose literacy and reading-writing requirements in 1994, about 2 million workers with low literacy skills were in jobs that presented them with few literacy requirements (Table 2). Close to 4 million had medium-level skills and were employed in jobs with mid-range requirements. Half of all workers with Level 2 prose literacy were in this situation, as were 55% of those in the next highest level, About 2.5 million Canadians with high literacy skills were in jobs requiring a high degree of prose literacy.

A reading-writing index and a numeracy index were used, to measure the fit or mismatch between. workers literacy skills and their job requirements. Both indices range from 1.0 to 5.0, because they were based on responses (numbered one to five) to the IALS questions. The values for each workplace requirement index were collapsed into four categories (1.0 to 1.99 = 1; 2.0 to 2.99 = 2; 3.0 to 3.99 = 3; 4.0 to 5.0 = 4) that reflected the range of categories ("rarely - never" to "every day") concerning workplace literacy requirements, with higher values indicating more frequent reading-writing or mathematical requirements.

These four-category measures were then cross-tabulated by the literacy (also four levels; Table 1) of employed sample members. Specifically, the distributions of prose literacy and document literacy were cross-tabulated by the reading-writing requirements measure (Tables 2 and 3, respectively), and quantitative literacy was cross-tabulated by the numeracy requirements measure (Table 4). For each literacy dimension examined in the IALS, five combinations were possible, given the construction of the two measures: low literacy skills and low literacy requirements in the workplace; medium literacy skills and medium literacy requirements; high literacy skills and high literacy requirements; low literacy skiffs and high literacy requirements (a literacy deficit); high literacy skills and low literacy requirements (a literacy surplus). Workers whose literacy skills roughly fit their job requirements (low-low, medium-medium and high-high) appear from top left to bottom right of the relevant table.

The tables also show the number who were mismatched; that is, those who exhibited either a literacy deficit (the upper right comer of the table) or a literacy surplus (the lower left corner of the table). The latter might also be described as "underemployed" in terms of their literacy skills. The deficit category includes workers whose measured literacy ability was at least two categories below the literacy requirement of their job. In contrast, those whose measured literacy ability was at least two categories above the literacy requirement of their job exhibited a surplus.