ECONOMIC DEVELOPMENT QUARTERLY / November 2001
Positive Negative Othera Total
8 1 8 17
Export and Employment Change Under NAFTA
Export Effects Negative
NOTE: NAFTA = North American Free Trade Agreement. Six of the nine cells have expected frequencies of less than 5, thus the chi-square value (45.06 at p < .05) is questionable. The 2 × 2 portion of the matrix (positive and negative export and job effects) is statistically significant (p = .054, Fisher’s exact test). a. Irrelevant, neutral, or impossible to tell.
NOTE: NAFTA = North American Free Trade Agreement. Twenty survey participants did not respond to questions about import competition.
Import competition from Canada
Incidence of NAFTA-Related Import Competition
Import Competition From Mexico No
Cheaper imports from Canada
NOTE: Thirty-nine survey respondents did not reply to questions about cheaper imports.
Incidence of Cheaper Input-Sourcing (imports)
Cheaper Imports From Mexico No
An implication here is that firms on a growth trajectory in terms of employment have also been on a growth path as far as exports are concerned.
(i.e., exports have increased both to Canada and Mexico). An implication here is that firms on a growth trajectory in terms of employment have also been on a growth path as far as exports are con- cerned. Data tabulated elsewhere indicate that the 8 firms that cited NAFTA in a positive light on the employment front are part of a distinct group of export-oriented firms (Will, 2000) in that most of these companies typically earn at least 30% of their annual sales from foreign markets.
This said, an obvious problem with Tables 1 to 5 is that we are dealing with categorical data rather than absolute numbers. Thus, the 5 firms indicating a negative employment impact (see Table 1) may have lost more jobs than the 8 firms indicating a positive impact (or vice versa). Pre- cisely the same criticism can be applied with regard to all of the other variables. In an effort to attach a sense of scale to the data, then, we conducted telephone interviews with two groups of firms. We defined the first group as positively affected (Group 1). This group consists of the 8 firms that indicated a positive employment impact. We defined the second group as negatively affected (Group 2). This group consists of the 5 firms that indicated a negative employment impact. Group 1 consists of 2 metal fabricators, 1 food processor, 3 producers of electronic goods, and