Part 1: Economic Theory of Minimum Wages

Standard economic theory of
minimum wages

It is a well-established fact that the quantity of a good or service demanded declines as its price rises relative to the prices of other goods and services. When the relative price of bananas rises, utility-maximizing consumers will buy fewer bananas and more oranges. Similarly, when the price of labour (i.e. the wage rate) rises, profit-maximizing firms will tend to substitute other inputs (e.g. machinery) for labour and reduce their demand for workers. This simple observation underscores the standard textbook conclusion that increases in minimum wages reduce employment.

In a competitive labour market, the demand for labour falls as real wages rise. Profit maximization on the part of the firm requires that the value of the marginal product of labour equal the real wage. Since, for a given level of capital and other inputs, each additional unit of labour produces less output, labour demand must vary inversely with the real wage. The effect of a minimum-wage law in this context is to reduce firms’ demand for labour. When the cost of labour input rises, the marginal product of labour must be higher than before if the firm is to minimize costs. However, because each additional unit of labour produces less output when other factors are fixed, the only way to increase the marginal product of labour is to reduce the number of workers employed.(In other words, labour is subject to diminishing returns.) Hence, the imposition of a minimum wage law will lead to fewer employment opportunities for those workers whose marginal productivity is below the minimum wage (Gunderson and Riddell 1988).

The extent to which an increase in the minimum wage reduces employment depends critically upon (1) the ability of firms to substitute other factors of production for labour in response to the change in relative input prices induced by the minimum wage increase; and (2) the impact the increase in the minimum wage has on total output of the firm and the industry. Suppose, on the one hand, that a firm’s production technology is such that production of a unit of output requires exactly one unit of capital and one unit of labour, and that no other combination of inputs can produce this unit of output; in other words, labour and capital are perfect complements in production.⁸ Under these circumstances, the firm cannot substitute capital for labour or labour for capital when relative input prices change. For a given level of output, an increase in the minimum wage will therefore have no disemployment effects. Of course, the increase in costs induced by the higher minimum wage will likely force some firms to reduce production or shut down completely. Less output in this context implies that fewer workers are employed. Job losses in this case result not from factor substitution per se but rather from the effects that the minimum wage has on total production. Suppose, on the other hand, that labour and capital are perfect substitutes in production. Under these circumstances, an increase in the relative price of labour will result in the firm substituting capital for labour completely and firing all its workers. In this scenario, an increase in the minimum wage will cause significant unemployment. Total output is unlikely to be affected much because capital is available as a perfect substitute for labour. Hence, job losses in this scenario result entirely from factor substitution on the part of the firm.

In reality, most production technologies fall somewhere in between the extremes of perfect substitute and perfect complement. Generally, capital and labour are imperfect substitutes for each other. In other words, there are several mixes of capital and labour that the firm can employ to produce a given amount of output. In these intermediate cases, the disemployment effects of high minimum wages are the product of both factor substitution and lower overall production. Hence, standard economic theory suggests that increases in the minimum wage should result in a fall in total employment.

The effects of minimum wages
on employment: the empirical
evidence

The bulk of the available empirical evidence on the effects of increases in the minimum wage on employment support the standard theory that increases in the minimum wage tend to reduce aggregate employment rates and increase unemployment rates.⁹ Furthermore, the young, women, the unskilled, and the working poor generally experience the disemployment effects of increases in the minimum wage. This is not surprising since these workers are generally among the least productive members of the work force and, hence, are most susceptible to increases in the minimum wage.

In a comprehensive survey of the empirical literature on minimum wages from the United States, Brown, Gilroy and Cohen (1982) conclude that the empirical evidence is generally in accord with the standard theory. Time-series econometric studies estimate that a 10 percent increase in the minimum wage reduces employment among teenagers (ages 15-19) by 1 percent to 3 percent, holding other factors constant. Empirical studies using cross-sectional data produce a wider range of estimates but most suggest that a 10 percent increase in the minimum wage reduces employment rates among teenagers up to 3 percent. Employment rates among young adults (ages 20-24) also decline with increases in minimum wages but the declines are smaller than those among teenagers.¹⁰ In a recent cross-country study, the Organisation for Economic Cooperation and Development (OECD) (1998) found that a 10 percent increase in the minimum wage rate reduced teenage employment rates by 2 percent to 4 percent. The impacts upon adult workers are more difficult to discern since higher minimum wages may induce firms to substitute adult labour for youth labour when minimum wages rise. Nonetheless, the impact of increases in the minimum wage upon total employment appears to be negative overall.

West and McKee (1980) survey the Canadian evidence and find that increases in the minimum wage reduce Canadian employment rates. Across most empirical studies, it seems that a 10 percent increase in the minimum wage reduces employment among Canadian teenagers by about 1 percent to 3 percent. Increases in the minimum wage also reduce employment rates among adults and young adults but not as dramatically as it reduces employment rates among teenagers (Schaafsma and Walsh 1983). Using more recent data, Law and Mihlar (1998) find that increases in provincial minimum wages reduce employment rates and raise unemployment rates among Canadian youth. Meanwhile, Shannon and Beach (1995) find that increases in Ontario’s minimum wage during the early 1990s reduced employment in retail sales and food services, and disproportionately affected young and part-time workers. Thus, the empirical evidence using Canadian data appears to be consistent with the standard economic theory of minimum wages.

Card and Krueger: the new twist to the minimum wage story

Recently, a number of analysts have come to question the standard view about the effects of minimum wages on employment. In particular, an influential study by economists David Card and Alan Krueger of Princeton University stands the consensus on its head. According to Card and Krueger (1994, 1995), the empirical evidence on the effects of increases in the minimum wage upon employment is not as convincing as the traditional analysis would suggest. They present evidence showing that increases in minimum wages in several American states did not result in employment losses and Card and Krueger claim that, in certain cases, increases in the minimum wage resulted in employment gains. For instance, in their much-cited study of employment in fast food restaurants in New Jersey and Pennsylvania (1994), they show that employment in New Jersey fast-food restaurants actually increased after the New Jersey legislature raised the minimum wage. These findings have sparked considerable controversy among labour economists and have generated much debate in academic and political circles (see, for instance, Ehrenburg 1995).

Card and Krueger justify their empirical results by arguing that in some instances, the labour market is a monopsony, i.e. a market with a single buyer. In principle, it is possible for an increase in the minimum wage to raise employment if a firm is a monopsonist (i.e. the sole purchaser) in the labour market (Gunderson and Riddell 1988). The classic example of monopsony in the labour market is a single-company town where there is only one major employer. Most analysts agree, however, that this theoretical argument is weak because very few labour markets are characterized by monopsony power, particularly over the long run when labour is mobile (Gunderson and Riddell 1988; Lal 1995). Hence, there does not appear to be a plausible basis for Card and Kreuger’s findings.

Moreover, many scholars are doubtful about the quality of Card and Kreuger’s empirical work. Rather than use traditional econometric techniques to estimate the employment effects of minimum wage increases, Card and Krueger use survey data to conduct “natural experiments” that do not control adequately for the various factors that affect the relevant variables. The particular methodology employed by Card and Krueger (1994) in their study of employment in fast-food restaurants in New Jersey and Pennsylvania has been criticized sharply by Daniel Hammermesh (1995), who argues that the timing of Card and Kreuger’s surveys bias their results. Finis Welch (1995), meanwhile, argues that by restricting their attention to fast-food restaurants, Card and Krueger neglect to consider the overall employment impacts of New Jersey’s increase in the minimum wage, which could very well have been negative. Finally, a study using official payroll data shows that employment did, in fact, fall in New Jersey following the increase in the minimum wage (Neumark and Wascher 1995a). Hence, there are a number of reasons to question the validity of Card and Kreuger’s conclusions.

Other impacts of minimum
wage laws

The preoccupation of many researchers with the employment effects of minimum wages obscures another important fact: that the labour market is more than just a spot market where money (i.e. wages) is exchanged for work. Mandated changes in the relative price of labour have effects beyond the employment rates. They will also affect the complex (but often unstated) contractual relationships within a firm even if employment is only affected negligibly. Suppose, for instance, that a firm’s production technology is characterized by perfect complementarity in capital and labour inputs and that an increase in the minimum wage has no disemployment effects because the firm’s output remains constant. Does this mean that everything else remains unchanged? Probably not. While it is true (by assumption) that the firm’s production technology prevents it from substituting capital for labour, it is also true that there are other margins of substitution available to the firm in its efforts to maximize profits (Oi 1997). If we view the labour market as more than just a spot market for work, then what matters to the firm is not strictly the real wage but rather the total compensation package that it offers to its workers. This total compensation package includes, in addition to wages, good working conditions, extra paid vacation time, flexible work hours, bonuses, and on-the-job training. An increase in the minimum wage will certainly increase the wage component of the total compensation package. However, because the firm’s concern is the total compensation package it offers to its workers and not strictly the wage component, it has every incentive to curtail other benefits when wages rise in order to contain total labour costs. Hence, even if total employment remains unchanged, an increase in the minimum wage may have other adverse consequences, including smaller bonuses, less on-the-job training, and fewer fringe benefits.

The available empirical evidence appears to support these intuitive arguments. Econometric studies from the early 1980s show that increases in minimum wages in the United States during the 1970s resulted in reduced expenditures on fringe benefits (Wessels, 1980). Empirical work by Hashimoto (1981) demonstrates that on-the-job training is reduced when minimum wages are increased. Since future earnings are known to be highly correlated with human capital, the dynamic consequences of reduced on-the-job training may be undesirable for young workers (Leighton and Mincer, 1981). Less on-the-job training reduces the formation of human capital and this means that future productivity is lower than it would otherwise be. Neumark and Wascher (1995b, 1996) show that high minimum wages also encourage young teenage workers to leave school in search of full-time employment. Because employment opportunities are scarcer when minimum wages are increased, teenage workers are generally unsuccessful in finding employment; the result is an increase in the number of idle teenagers, those neither in school nor employed. Hence, in addition to reducing employment rates, high minimum wages have a number of other subtle, yet significant, effects that are often ignored in policy debates.