Common menu bar

• Français
• Home
• Contact Us
• Help
• Search
• canada.gc.ca

Breadcrumb

1. Home >
2. Employment Standards >
3. Federal Labour Standards Review Commission

Archived - Minimum Wages in canada : theory, evidence and policy

Previous Table of Contents Next

Evidence From U.S. Studies

The empirical evidence based on the previously discussed methodologies are discussed based on evidence mainly from the United States since that is where the vast majority of the studies have been done. The review summarizes the impact of minimum wages on various aspects of labour market behaviour as outlined previously in the section on the theoretically expected impact of minimum wages. The next section discusses the British studies, followed by an analysis of the Canadian studies on a study-by-study basis since the number of studies makes this manageable, and a more detailed treatment is merited given their obvious relevance to policy making in Canada.

Impact on Employment

• Over the 1950s, 60s and 70s, the consensus, based mainly on time series studies was that a 10% increase in the minimum wage led to a 1% - 3% reduction in employment of teens[9], the group that was most often studied. That is, the elasticity of teenage employment with respect to the minimum wage was about –0.1 to –0.3. Almost all of the early studies found a negative relationship between minimum wages and employment, and most found the relationship to be statistically significant.
• This negative relationship was generally found for all age, sex, race subgroups, although it was largest for those whose wages were most likely to fall below the minimum (e.g., the youngest and Blacks)
• Importantly, the impacts tended to be smaller (slightly less than 1%) and less often statistically significant in the later studies[10] based on time series data from the 1980s as well as in the cross-section studies (across states or metropolitan areas), suggesting that the preferred estimates are likely to be at the lower end of the 1% to 3% range.
• The limited evidence[11] suggests that minimum wage increases also led to a slight reduction in hours of work, suggesting that the focus on employment tends to underestimate the total employment effect if the reduction in hours were also added to the employment effect.
• The limited evidence suggests that the reduction in employment gets translated into increased unemployment, but the effect is very small, implying that the minimum wage also leads to some labour force withdrawal (i.e., a slight negative effect on labour force participation).
• The impact on the employment of young adults (age 20-24) is also negative, but slightly smaller than that of teens suggesting that at least some of the adverse employment effect on young adults is offset by the positive employment effect emanating from the fact that some may be substitutes for teens when minimum wages are increased.[12]
• Few studies have examined the impact on adult employment and their results are inconclusive.
• The more recent studies that combine aggregate time series and cross section data across states, usually for the more recent period of the 1980s and into the 1990s, tend to find very conflicting results: sometimes within the former consensus range whereby a minimum wage increase of 10% would lead to an employment reduction of 1-3% (Neumark and Wascher 1992, 1994, Williams 1993, Williams and Mills 1998); even larger negative effects than the earlier consensus range (Burkhauser, Couch and Wittenburg 2000, Deere, Murphy and Welch 1995, Kim and Taylor 1995); smaller effects than the earlier consensus range, ranging from negative and insignificant (Card 1992a; Zavodny 2000), to positive but generally insignificant (Card and Krueger 1995) to positive and often significant (Card, Katz and Krueger 1994), to insignificant positive and negative depending upon gender and time period (Mills, Roy and Williams 1999).
• The studies that use panel data with individuals as the unit of observations to estimate the employment transitions of persons affected by minimum wage increases tend to find adverse employment effects that range from modest (Currie and Fallick 1996, Zavodny 2000) to substantial (Abowd, Kramarz, Lemieux and Margolis 2000, Linneman 1982) with Zavodny (2000) finding no effect on hours worked.
• Neumark (2001) is an important study because it uses recent pooled cross-section data for 1995-98, and, more importantly, it is based on a pre-specified research design discussed previously. Based on the results from 20 alternative specifications for five different groups, the following generalisations emerge:
  • There are never positive employment effects that are statistically significant.
  • Based on the groups that are used most often in the empirical literature (teens 16-19 and all youths 16-24) both positive and negative employment effects are found about equal numbers of times, but they are never statistically significant (hence being consistent with no employment effect)
  • When restricted to young persons who are not in school, and who have no more than high-school education, the employment effects are usually negative and significant about half of the time, albeit somewhat less so for young adults 20-24 when some specifications yield positive, albeit insignificant effects.
  • When adding the further restriction of not having completed high-school, the employment effect is always negative and significant in half of the specifications.
  • Overall, the results highlight the variability that occurs from alternative specifications and focusing on different groups, but seem most consistent with an interpretation of no substantial employment effects for teens and all youths, except for youths who are less educated and not in school
  • The more recent studies that use the “difference-in-difference” methodology based on natural experiments with before-and-after comparisons in “treatment” jurisdictions that increased their minimum wage compared to “control group” jurisdictions that did not increase their minimum wage, tend to find no adverse employment effect or even a positive employment effect that is sometimes statistically insignificant (Card 1992b, Katz and Krueger 1992, Card and Krueger 1994, 1995, 2000). Not surprisingly, given these findings, these studies have been hotly contested, and some “difference-in-difference” estimates do provide negative employment effects (Bellante and Picone 1999).
  • The exchange between Neumark and Wascher (2000) and Card and Krueger (2000) provide some resolution of the differences because they essentially re-examined the earlier natural experiment of the fast-food industry in New Jersey and Pennsylvania originally done in Card and Krueger (1994). Neumark and Wascher (2000) were critical of the fact that the original study was based on survey data of firms. They provided new evidence based on administrative payroll data of the firms and found that the survey data exhibited much more employment variability than did the payroll data (causing them to question the accuracy of the survey data). More importantly, they find that the payroll data imply that a 10% increase in the minimum wage would reduce employment by 1 - 2.5% which is almost exactly equal to the earlier consensus estimates of 1 - 3%, although their results are often statistically insignificant. They conclude (p. 1391) that

    “minimum wage increases reduced fast-food employment … [and] we can be more decisive in concluding that New Jersey’s minimum-wage increase did not raise fast-food employment in that state”.

    Card and Krueger (2000) respond to that critique by reanalysing the Neumark and Wascher payroll data (arguing that it is not based on a representative sample of employers) and by using two alternative sources of administrative payroll data provided by the Bureau of Labor Statistics. They conclude:

    “Consistent with our original sample, the BLS fast-food data set indicates slightly faster employment growth in New Jersey than in the Pennsylvania border counties over the time period that we initially examined, although in most specifications the differential is small and statistically insignificant (p. 1397) … the increase in the New Jersey minimum wage in April 1992 had little or no systematic effect on total fast food employment in the state (p. 1398) … The increase in New Jersey’s minimum wage probably had no effect on total employment in New Jersey’s fast-food industry, and possibly had a small positive effect” (p. 1419).

    One interpretation of this exchange is that both sides converged closer to zero effect. That is, although Neumark and Wascher (2000) found estimates that were in the earlier consensus range, they were often not statistically different from zero. As well, the new Card and Krueger (2000) estimates moved away from their earlier ones of often finding statistically significant positive employment effects, closer to ones that find no employment effect, and this seems their “preferred interpretation”.

  • Overall, it is extremely difficult to summarise the empirical evidence on the employment effect of minimum wages. Subject to that caveat, the following conclusions are offered:
  • The results vary by methodology, specification and group being examined
  • There is no consensus on the impact
  • The earlier consensus range of a 10% increase in the minimum wage leading to a 1 – 3% reduction in employment has likely widened to a “more fragile consensus” range of 0 – 3%. It is probably closer to the 0 end of the spectrum for most groups of teens and young adults, and possibly at the middle or higher end of the spectrum for less educated youths not in school, and in the longer run when firms have more opportunity to make the adjustments.
  • A positive employment impact is unlikely in the aggregate, albeit possible for some establishments.

Evidence on Labour Force Participation and Unemployment

Mincer (1976) and Ragan (1977) estimate separate equations (based on pooled cross section time series data) on the effect of minimum wages on teenage employment, labour force participation and unemployment, and they indirectly calculate the impact on unemployment from the employment and labour force participation equations. They find, in concordance with the general literature, that minimum wages reduce employment. They also find that minimum wages reduce labour force participation as people leave the labour force because of the fewer employment opportunities. The labour force withdrawal is not as large as the adverse employment effect; hence, unemployment increases. Alternatively stated, not all of the adverse employment effect gets translated into unemployment since many withdraw from the labour force, with the increase in unemployment being mitigated by the labour force withdrawal.

Evidence on Spillover or Ripple Effects

Although the number of studies is limited, the evidence on spillover is generally in agreement: small positive spillover effects exist, raising the wages of those slightly above the minimum wage, with the impact largely being confined to those just above the minimum[13].

Interestingly, this has both a “good news” and a “bad news” connotation. The good news is that this suggests that the minimum wage raises the wages of persons not only at the minimum wage, but also those just above the minimum wage. The bad news is that this could contaminate control group comparisons that rely on groups not directly affected by the minimum wage. If persons just above the minimum wage also have their wages increase for reasons of maintaining old relativities then they may also experience an adverse employment effect and hence the adverse employment effect of those affected directly by the minimum wage may be muted since they are being compared to a group whose employment is also declining because of their indirect wage increase. If their wages are increasing because of an increase in demand for their services reflecting a substitution away from minimum-wage workers, then this suggests that a small negative aggregate employment effect for a group, say teenagers, could be the net result of a large decline in the employment of persons affected directly by the minimum wage being partially offset by an expansion in the employment of persons just above the minimum wage. This “churning” would likely be regarded as undesirable even though it does not result in a large net employment reduction since otherwise fairly similar groups (e.g., low-wage teens just below and just above the minimum wage) receive very different outcomes. This violates the principle of horizontal equity whereby similar people should otherwise receive similar treatment.

Evidence on Permanency of Minimum Wage “Careers”

A limited number of empirical studies have documented that most minimum wage jobs for persons are temporary and that most people soon move out of such jobs into higher paying ones (Carrington and Fallick 2001, Long 1999, Schiller 1994, Smith and Vavrichek 1992). This is consistent with the notion that minimum wages disproportionately affect younger workers since such younger workers are most likely to move up the wage hierarchy as they gather experience.

While most minimum wage jobs are temporary and there is considerable upward mobility out of such jobs, Carrington and Fallick (2001, p. 17) “identify a nontrivial fraction of workers that spend substantial portions of their post-career on minimum wage or near minimum wage jobs. For example, we estimate that more than 8 percent of workers spend at least 50% of their first 10 post-school years working in jobs paying less than the minimum wage plus $1.00. We find that workers with such minimum wage careers are largely drawn from demographic groups with generally low wages: women, minorities and the less-educated. Thus, while relatively few in number, there is an identifiable subpopulation of workers whose lifetime income and employment is likely to be associated with minimum wages. For individuals in this group, minimum wages do not have merely transitory effects.” Of course, this does not imply that minimum wage laws would lift them out of such permanent minimum wage “careers” – they could move them out of employment altogether.

Previous Table of Contents Next

Primary navigation (left column)

Date Modified:

2011-11-22