Unemployment in Mexico : its characteristics and determinants

The restructuring of Mexico's economy has had surprisingly little effect on Mexican unemployment, which is low even in the worst years. The authors ask: Is the official definition of unemployment adequate? Is unemployment properly measured? And who bears its burden? Is the welfare cost of unemployment widespread or are certain population groups especially vulnerable to it, repeatedly hit by it, and therefore deserving of special attention? Is most unemployment associated with normal turnover (movements from one job to another) or with certain individuals being out of work a long time? The authors address these questions using panel data from the quarterly urban labor force survey, a household-based survey of 16 urban areas, and data from the National Employment Survey carried out every two to three years, and find the following. The structure of unemployment in Mexico is broadly similar to that in other countries. Unemployment is highest for those 16 to 25, especially women. Surprisingly, however, it is higher among secondary school graduates than among the less educated. Unemployment as officially measured is quite low, and has remained moderate during adjustment. Most adjustment occurred through the real wage rather than through unemployment. Therehas been relatively little restructuring for greater productivity. The official definition of unemployment leads to an underestimate of the jobless, because it ignores short spells out of the labor force transitions in and out of the labor force which are frequent. Using a more extension definition of unemployment raises the rate of male unemployment for 1988 from 3.4 percent to 6.4 percent, with the greatest increases in unemployment observed for people under 20 or with little education - yielding a structure of unemployment more like the one observed in other countries. Age, gender, and education are key determinants of unemployment. The probability of unemployment decreases with age and education for both men and women. Marriage is associated with lower risk of unemployment for men and for more educated women, but more probability of unemployment for women with less education. The typical spell of unemployment is not long: a mean duration of 5.7 months for men and 7.2 months for women (which explains the higher average unemployment rate for women). The duration of unemployment is longer for older workers but does not vary substantially according to educational attainment. Heads of households and individuals with household responsibilities tend to exit from unemployment faster. Although the typical spell of unemployment is relatively short, almost 70 percent of all unemployment in 1990-91 was attributable to spells lasting at least six months, and 30 percent corresponded to spells lasting at least a year. Although unemployment rates, as measured over a one-week period, are low (3 to 6 percent), 15 to 20 percent experience at least one spell of unemployment over a year. Among teenagers, the proportion is highest (50 percent) while it is only 10 percent for workers over 30.