Minimum wage

Customs and extra-legal pressures from governments or labor unions can produce a de facto minimum wage. So can international public opinion, by pressuring multinational companies to pay Third World workers wages usually found in more industrialized countries. The latter situation in Southeast Asia and Latin America was publicized in the 2000s, but it existed with companies in West Africa in the middle of the 20th century.[28]

Among the indicators that might be used to establish an initial minimum wage rate are ones that minimize the loss of jobs while preserving international competitiveness.[29] Among these are general economic conditions as measured by real and nominal gross domestic product; inflation; labor supply and demand; wage levels, distribution and differentials; employment terms; productivity growth; labor costs; business operating costs; the number and trend of bankruptcies; economic freedom rankings; standards of living and the prevailing average wage rate.

In the business sector, concerns include the expected increased cost of doing business, threats to profitability, rising levels of unemployment (and subsequent higher government expenditure on welfare benefits raising tax rates), and the possible knock-on effects to the wages of more experienced workers who might already be earning the new statutory minimum wage, or slightly more.[30] Among workers and their representatives, political considerations weigh in as labor leaders seek to win support by demanding the highest possible rate.[31] Other concerns include purchasing power, inflation indexing and standardized working hours.

In the United States, the minimum wage have been set under the Fair Labor Standards Act of 1938. According to the Economic Policy Institute, the minimum wage in the United States would have been $18.28 in 2013 if the minimum wage had kept pace with labor productivity.[32] To adjust for increased rates of worker productivity in the United States, raising the minimum wage to $22 (or more) an hour has been presented.[33][34][35][36]

Graph showing the basic supply and demand model of the minimum wage in the labor market.

According to the supply and demand model of the labor market shown in many economics textbooks, increasing the minimum wage decreases the employment of minimum-wage workers.[8] One such textbook states:[4]

If a higher minimum wage increases the wage rates of unskilled workers above the level that would be established by market forces, the quantity of unskilled workers employed will fall. The minimum wage will price the services of the least productive (and therefore lowest-wage) workers out of the market. … the direct results of minimum wage legislation are clearly mixed. Some workers, most likely those whose previous wages were closest to the minimum, will enjoy higher wages. Others, particularly those with the lowest prelegislation wage rates, will be unable to find work. They will be pushed into the ranks of the unemployed.

A firm's cost is an increasing function of the wage rate. The higher the wage rate, the fewer hours an employer will demand of employees. This is because, as the wage rate rises, it becomes more expensive for firms to hire workers and so firms hire fewer workers (or hire them for fewer hours). The demand of labor curve is therefore shown as a line moving down and to the right.[37] Since higher wages increase the quantity supplied, the supply of labor curve is upward sloping, and is shown as a line moving up and to the right.[37] If no minimum wage is in place, wages will adjust until quantity of labor demanded is equal to quantity supplied, reaching equilibrium, where the supply and demand curves intersect. Minimum wage behaves as a classical price floor on labor. Standard theory says that, if set above the equilibrium price, more labor will be willing to be provided by workers than will be demanded by employers, creating a surplus of labor, i.e. unemployment.[37] The economic model of markets predicts the same of other commodities (like milk and wheat, for example): Artificially raising the price of the commodity tends to cause an increase in quantity supplied and a decrease in quantity demanded. The result is a surplus of the commodity. When there is a wheat surplus, the government buys it. Since the government does not hire surplus labor, the labor surplus takes the form of unemployment, which tends to be higher with minimum wage laws than without them.[28]

The supply and demand model implies that by mandating a price floor above the equilibrium wage, minimum wage laws will cause unemployment.[38][39] This is because a greater number of people are willing to work at the higher wage while a smaller number of jobs will be available at the higher wage. Companies can be more selective in those whom they employ thus the least skilled and least experienced will typically be excluded. An imposition or increase of a minimum wage will generally only affect employment in the low-skill labor market, as the equilibrium wage is already at or below the minimum wage, whereas in higher skill labor markets the equilibrium wage is too high for a change in minimum wage to affect employment.[40]

Modern economics suggests that a moderate minimum wage may increase employment as labor markets are monopsonistic and workers lack bargaining power.

The supply and demand model predicts that raising the minimum wage helps workers whose wages are raised, and hurts people who are not hired (or lose their jobs) when companies cut back on employment. But proponents of the minimum wage hold that the situation is much more complicated than the model can account for. One complicating factor is possible monopsony in the labor market, whereby the individual employer has some market power in determining wages paid. Thus it is at least theoretically possible that the minimum wage may boost employment. Though single employer market power is unlikely to exist in most labor markets in the sense of the traditional 'company town,' asymmetric information, imperfect mobility, and the personal element of the labor transaction give some degree of wage-setting power to most firms.[41]

Modern economic theory predicts that although an excessive minimum wage may raise unemployment as it fixes a price above most demand for labor, a minimum wage at a more reasonable level can increase employment, and enhance growth and efficiency. This is because labor markets are monopsonistic and workers persistently lack bargaining power. When poorer workers have more to spend it stimulates effective aggregate demand for goods and services.[42][43]

The argument that a minimum wage decreases employment is based on a simple supply and demand model of the labor market. A number of economists (for example Pierangelo Garegnani,[44] Robert L. Vienneau,[45] and Arrigo Opocher & Ian Steedman[46]), building on the work of Piero Sraffa, argue that that model, even given all its assumptions, is logically incoherent. Michael Anyadike-Danes and Wynne Godley[47] argue, based on simulation results, that little of the empirical work done with the textbook model constitutes a potentially falsifiable theory, and consequently empirical evidence hardly exists for that model. Graham White[48] argues, partially on the basis of Sraffianism, that the policy of increased labor market flexibility, including the reduction of minimum wages, does not have an "intellectually coherent" argument in economic theory.

Minimum wage levels in OECD countries as a share of average full-time wage, 2013.[49]

Gary Fields, Professor of Labor Economics and Economics at Cornell University, argues that the standard textbook model for the minimum wage is ambiguous, and that the standard theoretical arguments incorrectly measure only a one-sector market. Fields says a two-sector market, where "the self-employed, service workers, and farm workers are typically excluded from minimum-wage coverage... [and with] one sector with minimum-wage coverage and the other without it [and possible mobility between the two]," is the basis for better analysis. Through this model, Fields shows the typical theoretical argument to be ambiguous and says "the predictions derived from the textbook model definitely do not carry over to the two-sector case. Therefore, since a non-covered sector exists nearly everywhere, the predictions of the textbook model simply cannot be relied on."[50]

An alternate view of the labor market has low-wage labor markets characterized as monopsonistic competition wherein buyers (employers) have significantly more market power than do sellers (workers). This monopsony could be a result of intentional collusion between employers, or naturalistic factors such as segmented markets, search costs, information costs, imperfect mobility and the personal element of labor markets. In such a case a simple supply and demand graph would not yield the quantity of labor clearing and the wage rate. This is because while the upward sloping aggregate labor supply would remain unchanged, instead of using the upward labor supply curve shown in a supply and demand diagram, monopsonistic employers would use a steeper upward sloping curve corresponding to marginal expenditures to yield the intersection with the supply curve resulting in a wage rate lower than would be the case under competition. Also, the amount of labor sold would also be lower than the competitive optimal allocation.

Such a case is a type of market failure and results in workers being paid less than their marginal value. Under the monopsonistic assumption, an appropriately set minimum wage could increase both wages and employment, with the optimal level being equal to the marginal product of labor.[51] This view emphasizes the role of minimum wages as a market regulation policy akin to antitrust policies, as opposed to an illusory "free lunch" for low-wage workers.

Another reason minimum wage may not affect employment in certain industries is that the demand for the product the employees produce is highly inelastic.[52] For example, if management is forced to increase wages, management can pass on the increase in wage to consumers in the form of higher prices. Since demand for the product is highly inelastic, consumers continue to buy the product at the higher price and so the manager is not forced to lay off workers. Economist Paul Krugman argues this explanation neglects to explain why the firm was not charging this higher price absent the minimum wage.[53]

Three other possible reasons minimum wages do not affect employment were suggested by Alan Blinder: higher wages may reduce turnover, and hence training costs; raising the minimum wage may "render moot" the potential problem of recruiting workers at a higher wage than current workers; and minimum wage workers might represent such a small proportion of a business's cost that the increase is too small to matter. He admits that he does not know if these are correct, but argues that "the list demonstrates that one can accept the new empirical findings and still be a card-carrying economist."[54]

The following mathematical models are more quantitative in orientation, and highlight some of the difficulties in determining the impact of the minimum wage on labor market outcomes.[55] Specifically, these models focus on labor markets with frictions.

Assume that the decision to participate in the labor market results from a trade-off between being an unemployed job seeker and not participating at all. All individuals whose expected utility outside the labor market is less than the expected utility of an unemployed person ${\displaystyle V_{u}}$ decide to participate in the labor market. In the basic search and matching model, the expected utility of unemployed persons ${\displaystyle V_{u}}$ and that of employed persons ${\displaystyle V_{e}}$ are defined by:

$${\displaystyle {\begin{aligned}rV_{e}&=w+q(V_{u}-V_{e})\\rV_{u}&=z+\theta m(\theta )(V_{e}-V_{u})\end{aligned}}}$$

Let ${\displaystyle w}$ be the wage, ${\displaystyle r}$ the interest rate, ${\displaystyle z}$ the instantaneous income of unemployed persons, ${\displaystyle q}$ the exogenous job destruction rate, ${\displaystyle \theta }$ the labor market tightness, and ${\displaystyle \theta m(\theta )}$ the job finding rate. The profits ${\displaystyle \Pi _{e}}$ and ${\displaystyle \Pi _{v}}$ expected from a filled job and a vacant one are:

$${\displaystyle r\Pi _{e}=y-w+q(\Pi _{v}-\Pi _{e}),\quad r\Pi _{v}=-h+m(\theta )(\Pi _{e}-\Pi _{v})}$$

where ${\displaystyle h}$ is the cost of a vacant job and ${\displaystyle y}$ is the productivity. When the free entry condition ${\displaystyle \Pi _{v}=0}$ is satisfied, these two equalities yield the following relationship between the wage ${\displaystyle w}$ and labor market tightness ${\displaystyle \theta }$:

Part of a series on
Organized labor
Labor movement Timeline New Unionism Proletariat Social movement unionism Social democracy Socialism Syndicalism Anarcho-syndicalism
Labor rights Freedom of association Collective bargaining Child labor Unfree labour Equal pay Employment discrimination Legal working age Eight-hour day Annual leave Minimum wage Occupational safety and health Employment protection
Trade unions Trade unions by country Trade union federations International comparisons ITUC IWA WFTU
Strike action Chronological list of strikes General strike Secondary action Sitdown strike Work-to-rule
Labor parties Australia Britain Ireland Israel Malta New Zealand Netherlands Norway List of Labor parties
Academic disciplines Industrial relations Labor economics Labor history Labor law

$${\displaystyle {h \over {m(\theta )}}={y-w \over {r+q}}}$$

If ${\displaystyle w}$ represents a minimum wage that applies to all workers, this equation completely determines the equilibrium value of the labor market tightness ${\displaystyle \theta }$. There are two conditions associated with the matching function:

$${\displaystyle m'(\theta )<0,\quad [\theta m(\theta )]'>0}$$

This implies that ${\displaystyle \theta }$ is a decreasing function of the minimum wage ${\displaystyle w}$, and so is the job finding rate ${\displaystyle \alpha =\theta m(\theta )}$. A hike in the minimum wage degrades the profitability of a job, so firms post fewer vacancies and the job finding rate falls off. Now let's rewrite ${\displaystyle rV_{u}}$ to be:

$${\displaystyle rV_{u}={(r+q)z+\theta m(\theta )w \over {r+q+\theta m(\theta )}}}$$

Using the relationship between the wage and labor market tightness to eliminate the wage from the last equation gives us:

$${\displaystyle rV_{u}={\theta m(\theta )y+(r+q)z-\theta (r+q)h \over {r+q+\theta m(\theta )}}}$$

If we maximize ${\displaystyle rV_{u}}$ in this equation, with respect to the labor market tightness, we find that:

$${\displaystyle {[1-\eta (\theta )](y-z) \over {r+q+\eta (\theta )\theta m(\theta )}}={h \over {m(\theta )}}}$$

where ${\displaystyle \eta (\theta )}$ is the elasticity of the matching function:

$${\displaystyle \eta (\theta )=-\theta {m'(\theta ) \over {m(\theta )}}\equiv -\theta {d \over {d\theta }}\log m(\theta )}$$

This result shows that the expected utility of unemployed workers is maximized when the minimum wage is set at a level that corresponds to the wage level of the decentralized economy in which the bargaining power parameter is equal to the elasticity ${\displaystyle \eta (\theta )}$.  The level of the negotiated wage is ${\displaystyle w^{*}}$.

If ${\displaystyle w<w^{*}}$, then an increase in the minimum wage increases participation and the unemployment rate, with an ambiguous impact on employment. When the bargaining power of workers is less than ${\displaystyle \eta (\theta )}$, an increases in the minimum wage improves the welfare of the unemployed - this suggests that minimum wage hikes can improve labor market efficiency, at least up to the point when bargaining power equals ${\displaystyle \eta (\theta )}$. On the other hand, if ${\displaystyle w\geq w^{*}}$, any increases in the minimum wage entails a decline in labor market participation and an increase in unemployment.

In the model just presented, we found that the minimum wage always increases unemployment. This result does not necessarily hold when the search effort of workers in endogenous.

Consider a model where the intensity of the job search is designated by the scalar ${\displaystyle \epsilon }$, which can be interpreted as the amount of time and/or intensity of the effort devoted to search. Assume that the arrival rate of job offers is ${\displaystyle \alpha \epsilon }$ and that the wage distribution is degenerated to a single wage ${\displaystyle w}$. Denote ${\displaystyle \varphi (\epsilon )}$ to be the cost arising from the search effort, with ${\displaystyle \varphi '>0,\;\varphi ''>0}$. Then the discounted utilities are given by:

$${\displaystyle {\begin{aligned}rV_{e}&=w+q(V_{u}-V_{e})\\rV_{u}&=\max _{\epsilon }\;z-\varphi (\epsilon )+\alpha \epsilon (V_{e}-V_{u})\end{aligned}}}$$

Therefore, the optimal search effort is such that the marginal cost of performing the search is equation to the marginal return:

$${\displaystyle \varphi '(\epsilon )=\alpha (V_{e}-V_{u})}$$

This implies that the optimal search effort increases as the difference between the expected utility of the job holder and the expected utility of the job seeker grows. In fact, this difference actually grows with the wage. To see this, take the difference of the two discounted utilities to find:

$${\displaystyle (r+q)(V_{e}-V_{u})=w-\max _{\epsilon }\left[z-\varphi (\epsilon )+\alpha \epsilon (V_{e}-V_{u})\right]}$$

Then differentiating with respect to ${\displaystyle w}$ and rearranging gives us:

$${\displaystyle {d \over {dw}}(V_{e}-V_{u})={1 \over {r+q+\alpha \epsilon ^{*}}}>0}$$

where ${\displaystyle \epsilon ^{*}}$ is the optimal search effort. This implies that a wage increase drives up job search effort and, therefore, the job finding rate. Additionally, the unemployment rate ${\displaystyle u}$ at equilibrium is given by:

$${\displaystyle u={q \over {q+\alpha \epsilon }}}$$

A hike in the wage, which increases the search effort and the job finding rate, decreases the unemployment rate. So it is possible that a hike in the minimum wage may, by boosting the search effort of job seekers, boost employment. Taken in sum with the previous section, the minimum wage in labor markets with frictions can improve employment and decrease the unemployment rate when it is sufficiently low. However, a high minimum wage is detrimental to employment and increases the unemployment rate.

In 1992, the minimum wage in New Jersey increased from $4.25 to $5.05 per hour (an 18.8% increase), while in the adjacent state of Pennsylvania it remained at $4.25. David Card and Alan Krueger gathered information on fast food restaurants in New Jersey and eastern Pennsylvania in an attempt to see what effect this increase had on employment within New Jersey. A basic supply and demand model predicts that relative employment should have decreased in New Jersey. Card and Krueger surveyed employers before the April 1992 New Jersey increase, and again in November–December 1992, asking managers for data on the full-time equivalent staff level of their restaurants both times.[72] Based on data from the employers' responses, the authors concluded that the increase in the minimum wage slightly increased employment in the New Jersey restaurants.[72]

Card and Krueger expanded on this initial article in their 1995 book Myth and Measurement: The New Economics of the Minimum Wage.[73] They argued that the negative employment effects of minimum wage laws are minimal if not non-existent. For example, they look at the 1992 increase in New Jersey's minimum wage, the 1988 rise in California's minimum wage, and the 1990–91 increases in the federal minimum wage. In addition to their own findings, they reanalyzed earlier studies with updated data, generally finding that the older results of a negative employment effect did not hold up in the larger datasets.[74]

A 2010 study published in the Review of Economics and Statistics compared 288 pairs of contiguous U.S. counties with minimum wage differentials from 1990 to 2006 and found no adverse employment effects from a minimum wage increase. Contiguous counties with different minimum wages are in purple. All other counties are in white.[75]

In 1996, David Neumark and William Wascher reexamined Card and Krueger's result using administrative payroll records from a sample of large fast food restaurant chains, and reported that minimum wage increases were followed by decreases in employment. An assessment of data collected and analyzed by Neumark and Wascher did not initially contradict the Card and Krueger results,[76] but in a later edited version they found a four percent decrease in employment, and reported that "the estimated disemployment effects in the payroll data are often statistically significant at the 5- or 10-percent level although there are some estimators and subsamples that yield insignificant—although almost always negative" employment effects.[77] Neumark and Wascher's conclusions were subsequently rebutted in a 2000 paper by Card and Krueger.[78] A 2011 paper has reconciled the difference between Card and Krueger's survey data and Neumark and Wascher's payroll-based data. The paper shows that both datasets evidence conditional employment effects that are positive for small restaurants, but are negative for large fast-food restaurants.[79] A 2014 analysis based on panel data found that the minimum wage reduces employment among teenagers.[80]

In 1996 and 1997, the federal minimum wage was increased from $4.25 to $5.15, thereby increasing the minimum wage by $0.90 in Pennsylvania but by just $0.10 in New Jersey; this allowed for an examination of the effects of minimum wage increases in the same area, subsequent to the 1992 change studied by Card and Krueger. A study by Hoffman and Trace found the result anticipated by traditional theory: a detrimental effect on employment.[81]

Further application of the methodology used by Card and Krueger by other researchers yielded results similar to their original findings, across additional data sets.[82] A 2010 study by three economists (Arindrajit Dube of the University of Massachusetts Amherst, William Lester of the University of North Carolina at Chapel Hill, and Michael Reich of the University of California, Berkeley), compared adjacent counties in different states where the minimum wage had been raised in one of the states. They analyzed employment trends for several categories of low-wage workers from 1990 to 2006 and found that increases in minimum wages had no negative effects on low-wage employment and successfully increased the income of workers in food services and retail employment, as well as the narrower category of workers in restaurants.[82][83]

However, a 2011 study by Baskaya and Rubinstein of Brown University found that at the federal level, "a rise in minimum wage have [sic] an instantaneous impact on wage rates and a corresponding negative impact on employment", stating, "Minimum wage increases boost teenage wage rates and reduce teenage employment."[84] Another 2011 study by Sen, Rybczynski, and Van De Waal found that "a 10% increase in the minimum wage is significantly correlated with a 3−5% drop in teen employment."[85] A 2012 study by Sabia, Hansen, and Burkhauser found that "minimum wage increases can have substantial adverse labor demand effects for low-skilled individuals", with the largest effects on those aged 16 to 24.[86]

A 2013 study by Meer and West concluded that "the minimum wage reduces net job growth, primarily through its effect on job creation by expanding establishments ... most pronounced for younger workers and in industries with a higher proportion of low-wage workers."[87] This study by Meer and West was later critiqued for its trends of assumption in the context of narrowly defined low-wage groups.[88] The authors replied to the critiques and released additional data which addressed the criticism of their methodology, but did not resolve the issue of whether their data showed a causal relationship.[89][90] A 2019 paper published in the Quarterly Journal of Economics by Cengiz, Dube, Lindner and Zipperer argues that the job losses found using a Meer and West type methodology "tend to be driven by an unrealistically large drop in the number of jobs at the upper tail of the wage distribution, which is unlikely to be a causal effect of the minimum wage."[91] Another 2013 study by Suzana Laporšek of the University of Primorska, on youth unemployment in Europe claimed there was "a negative, statistically significant impact of minimum wage on youth employment."[92] A 2013 study by labor economists Tony Fang and Carl Lin which studied minimum wages and employment in China, found that "minimum wage changes have significant adverse effects on employment in the Eastern and Central regions of China, and result in disemployment for females, young adults, and low-skilled workers".[93][94]

A 2017 study found that in Seattle, increasing the minimum wage to $13 per hour lowered income of low-wage workers by $125 per month, due to the resulting reduction in hours worked, as industries made changes to make their businesses less labor intensive. The authors argue that previous research that found no negative effects on hours worked are flawed because they only look at select industries, or only look at teenagers, instead of entire economies.[95]

Finally, a study by Overstreet in 2019 examined increases to the minimum wage in Arizona. Utilizing data spanning from 1976 to 2017, Overstreet found that a 1% increase in the minimum wage was significantly correlated with a 1.13% increase in per capita income in Arizona. This study could show that smaller increases in minimum wage may not distort labor market as significantly as larger increases experienced in other cities and states. Thus, the small increases experienced in Arizona may have actually led to a slight increase in economic growth.[96]

In 2019, economists from Georgia Tech published a study that found a strong correlation between incresases to the minimum wage and detectable harm to the financial conditions of small businesses, including a higher rate of bankruptcy, lower hiring rates, lower credit scores, and higher interest payments. The researchers noted that these small businesses were also correlated with minority ownership and minority customer bases. [97]

In July 2019, the Congressional Budget Office published the impact on proposed national $15/hour legislation. It noted that workers who retained full employment would see a modest improvement in take home pay offset by a small decrease in working conditions and non-pecuniary benefits. However, this benefit is offset by three primary factors; the reduction in hours worked, the reduction in total employment, and the increased cost of goods and services. Those factors result in a decrease of about $33 Billion in total income and nearly 1.7-3.7 million lost jobs in the first three years (the CBO also noted this figure increases over time). [98]

In response to an April 2016 Council of Economic Advisers (CEA) report advocating the raising of the minimum wage to deter crime, economists used data from the 1998-2016 Uniform Crime Reports (UCR), National Incident-Based Reporting System (NIBRS), and National Longitudinal Study of Youth (NLSY) to assess the impact of the minimum wage on crime. They found that increasing the minimum wage resulted in increased property crime arrests among those ages 16-to-24. They estimated that an increase of the Federal minimum wage to $15/hour would "generate criminal externality costs of nearly $2.4 billion." [99]

Economists in Denmark, relying on a discontinuity in wage rates when a worker turns 18, found that employment fell by 33% and total hours fell by 45% when the minimum wage law was in effect. [100]

Several researchers have conducted statistical meta-analyses of the employment effects of the minimum wage. In 1995, Card and Krueger analyzed 14 earlier time-series studies on minimum wages and concluded that there was clear evidence of publication bias (in favor of studies that found a statistically significant negative employment effect). They point out that later studies, which had more data and lower standard errors, did not show the expected increase in t-statistic (almost all the studies had a t-statistic of about two, just above the level of statistical significance at the .05 level).[101] Though a serious methodological indictment, opponents of the minimum wage largely ignored this issue; as Thomas Leonard noted, "The silence is fairly deafening."[102]

In 2005, T.D. Stanley showed that Card and Krueger's results could signify either publication bias or the absence of a minimum wage effect. However, using a different methodology, Stanley concluded that there is evidence of publication bias and that correction of this bias shows no relationship between the minimum wage and unemployment.[103] In 2008, Hristos Doucouliagos and T.D. Stanley conducted a similar meta-analysis of 64 U.S. studies on disemployment effects and concluded that Card and Krueger's initial claim of publication bias is still correct. Moreover, they concluded, "Once this publication selection is corrected, little or no evidence of a negative association between minimum wages and employment remains."[104] In 2013, a meta-analysis of 16 UK studies found no significant effects on employment attributable to the minimum wage.[105] a 2007 meta-analyses by David Neumark of 96 studies found a consistent, but not always statistically significant, negative effect on employment from increases in the minimum wage.[106]

A basic income (or negative income tax - NIT) is a system of social security that periodically provides each citizen with a sum of money that is sufficient to live on frugally. Supporters of the basic-income idea argue that recipients of the basic income would have considerably more bargaining power when negotiating a wage with an employer, as there would be no risk of destitution for not taking the employment. As a result, jobseekers could spend more time looking for a more appropriate or satisfying job, or they could wait until a higher-paying job appeared. Alternatively, they could spend more time increasing their skills (via education and training), which would make them more suitable for higher-paying jobs, as well as provide numerous other benefits. Experiments on Basic Income and NIT in Canada and the USA show that people spent more time studying while the program was running.[175]

Proponents argue that a basic income that is based on a broad tax base would be more economically efficient than a minimum wage, as the minimum wage effectively imposes a high marginal tax on employers, causing losses in efficiency.

A guaranteed minimum income is another proposed system of social welfare provision. It is similar to a basic income or negative income tax system, except that it is normally conditional and subject to a means test. Some proposals also stipulate a willingness to participate in the labor market, or a willingness to perform community services.[176]

A refundable tax credit is a mechanism whereby the tax system can reduce the tax owed by a household to below zero, and result in a net payment to the taxpayer beyond their own payments into the tax system. Examples of refundable tax credits include the earned income tax credit and the additional child tax credit in the US, and working tax credits and child tax credits in the UK. Such a system is slightly different from a negative income tax, in that the refundable tax credit is usually only paid to households that have earned at least some income. This policy is more targeted against poverty than the minimum wage, because it avoids subsidizing low-income workers who are supported by high-income households (for example, teenagers still living with their parents).[177]

In the United States, earned income tax credit rates, also known as EITC or EIC, vary by state—some are refundable while other states do not allow a refundable tax credit.[178] The federal EITC program has been expanded by a number of presidents including Jimmy Carter, Ronald Reagan, George H.W. Bush, and Bill Clinton.[179] In 1986, President Reagan described the EITC as "the best anti poverty, the best pro-family, the best job creation measure to come out of Congress."[180] The ability of the earned income tax credit to deliver larger monetary benefits to the poor workers than an increase in the minimum wage and at a lower cost to society was documented in a 2007 report by the Congressional Budget Office.[181]

The Adam Smith Institute prefers cutting taxes on the poor and middle class instead of raising wages as an alternative to the minimum wage.[182]

Italy, Sweden, Norway, Finland, and Denmark are examples of developed nations where there is no minimum wage that is required by legislation.[20][22] Such nations, particularly the Nordics, have very high union participation rates.[183] Instead, minimum wage standards in different sectors are set by collective bargaining.[184]

Some economists such as Scott Sumner[185] and Edmund Phelps[186] advocate a wage subsidy program. A wage subsidy is a payment made by a government for work people do. It is based either on an hourly basis or by income earned. Advocates argue that the primary deficiencies of the EITC and the minimum wage are best avoided by a wage subsidy.[187][188] However, the wage subsidy in the United States suffers from a lack of political support from either major political party.[189][190]

Providing education or funding apprenticeships or technical training can provide a bridge for low skilled workers to move into wages above a minimum wage. For example, Germany has adopted a state funded apprenticeship program that combines on-the-job and classroom training.[191] Having more skills makes workers more valuable and more productive, but having a high minimum wage for low-skill jobs reduces the incentive to seek education and training.[192] Moving some workers to higher-paying jobs will decrease the supply of workers willing to accept low-skill jobs, increasing the market wage for those low skilled jobs (assuming a stable labor market). However, in that solution the wage will still not increase above the marginal return for the role and will likely promote automation or business closure.