Résumés
Résumé
Les principales conclusions de cet article sont qu'un niveau élevé du salaire minimum, sans poser d'obstacle insurmontable quant à la stabilité des prix dans une économie aussi exposée à la concurrence extérieure que l'économie du Québec, risque de soutenir un niveau de chômage plus important qu'on a pu le croire jusqu'à ce jour, et tout spécialement parmi les jeunes; et que le salaire minimum élevé est un moyen très inefficace de mieux répartir les revenus qu'il vaudrait mieux remplacer par des outils d'intervention mieux appropriés comme un régime de revenu familial garanti ou même des plans de soutien du revenu un peu plus modestes.
Abstract
From 1972 to 1976, the Bourassa government raised the ratio of minimum to average industrial wages in Québec from 44 to 51%. The Lévesque administration maintained the minimum wage at that relative level at least to 1978. Thus, in the 1970s, the Québec minimum wage became the highest in North America.
This paper provides a summary of a report on the effects of the minimum wage written by the author for the Québec Government in the Spring of 1978. The four-step study attempted to:
1) evaluate the probable impact of the minimum wage on other wages and prices;
2) estimate the disemployment effect of the regulation;
3) compare the income-redistributive effect of the minimum wage (as a result of the first two steps) with the family income needs of the low-wage workers;
4) examine the interaction of the minimum wage with social assistance, and the alternatives offered by guaranteed family income plans and wage subsidies.
An increase of 10 per cent of the relative minimum wage was estimated to raise the overall industrial wage bill directly by about 0.4 per cent, and indirectly by another 0.2 to 0.3 per cent through wage emulation. Given the weight of the wage bill in total costs of production, the potential inflationary effect of the measure on the consumer price index was estimated to be moderate (0.3 and 0.5 per cent). The incidence of the increase of minimum wage varies of course widely across industries, depending on their relative use of low-wage manpower. Moreover, theactual increase in the CPI may be even lower than 0.3 per cent if competition in the very open Québec economy induces firms to adjust more through reductions in employment than through passing the buck to the consumer.
The disemployment effects of the minimum wage were found to be quantitatively important. After taking into account cyclical factors, demographic changes, and the 1971 reform of the Unemployment Insurance Act, it was found that a 10 per cent increase in the ratio of minimum to average industry wage raised the unemployment rate of young men aged 15 to 24 by 2.5 to 3.5 percentage points, that of young women aged 15 to 24 by 1.5 to 3.0 percentage points, and that of adult women aged 25 and over by 0.4 to 0.7 percentage points. In turn, this resulted in an upward effect of 0.6 to 1.0 percentage point on the overall unemployment rate (assuming the unemployment rate of adult men aged 25 and over remained unaffected).
Four reasons are suggested for the significant size for these effects: 1) the minimum wage law covers close to 100 per cent of non-farm workers; 2) the disemployment effect is highly nonlinear, i.e. it is more important if the minimum wage is initially high than if it is initially low; 3) in an economy as open as that of Québec, firms adjust to a higher minimum wage more through employment reductions than through price increases; and 4) generous unemployment compensation in Canada weakens the "discouraged worker effect" and exacerbates the measured unemployment consequences of any minimum wage increase.
The disemployment effects of the Québec minimum wage are so important in some categories of (mainly young) workers that anincrease in the relative minimum wage often results in adecrease of annual earned income. It was estimated that a 10 per cent minimum wage hike would, on average, change the earned incomes of low-wage workers as follows: a decrease of up to 13 per cent for young men, a decrease of 2 per cent to an increase of 4 per cent for young women, and an increase of 4 to 6 per cent for adult women. However, if one notes that unemployment insurance benefits are able to cushion up to 75 per cent of the net wage loss from unemployment, thesum of earned income and UI benefits almost always increases after a rise in the minimum wage. In other words, the reason why a high relative minimum wage helps many workers is because the Unemployment Insurance Commission takes charge of a larger fraction of their total incomes.
Moreover, there is evidence to show that over 80 per cent of minimum-wage workers are either young persons (mostly living with their parents), unattached individuals or the second wage-earners of childless families, all of whose family incomes are typically 50 to 100 per cent higher than Statistics Canada's "low income levels" (even before property income and tips are accounted for). The general picture is therefore that of an indiscriminate measure which often does not help the average worker to earn a higher income and whose incidence falls mainly on workers with family incomes well beyond poverty levels.
The minimum wage interacts strongly with social assistance (SA) benefits. The Québec Government tends to index the former to the cost of living because the latter are indexed by statute, and because the net annual income from full-time work at the minimum wage is already smaller than or equal to the scale of SA benefits in all kinds of families except the one-adult-no-children families. It was shown in the report, however, that a decrease in the ratio of the minimum wage to SA benefits would generate a work disincentive effect and slight increase in SA rolls, more than offset, however, by the reemployment effect of the lower minimum wage and by the implied decrease in UI rolls. However, since this would create a net cost for the Québec Government (who is responsible for the minimum wage regulation and social assistance) and a (much larger) net saving for the federal government (who controls unemployment insurance), there is no great incentive at this time for the Québec government to adopt such a measure.
A guaranteed family income plan (GFIP) is suggested as an alternative. Abolition of the minimum wage under a GFIP would not necessarily have the overall costs of social security skyrocket, because the resulting increase in average weekly hours of work would moderate the slow-down in weekly wages, and because the reemployment effect would lower unemployment insurance costs significantly. All depends on the GFIP tax rate. However, implicit intergovernmental transfers again enter the picture. And the very high cost of a GFIP may lead one to revert to more modest alternatives (targeted wage subsidies or work income supplements, public employment, etc.).
All in all, it was suggested that the Québec minimum wage regulation should be maintained, but that the rate be brought more in line with North American standards, and that more selective measures to help the true working poor replace the high minimum wage.