Public-sector unions are central to some of the most contentious public policy issues in contemporary America. Do unions improve or impair the political process and elections? Do police unions protect bad cops or support a beleaguered profession? Do we attribute the stagnation of the nation’s test scores to union protection of incompetent teachers or insufficient resources? Has union influence imperiled state and local budgets or provided a means of social mobility through public service?

In sum, would America’s political, economic, and social fortunes improve with more union influence or less? What about North Carolina’s?

This study will focus on state-level fiscal effects of the repeal of North Carolina’s statutory prohibition on collective bargaining for public-sector employees. The author argues that a repeal of the law would lead to sizable increases in state government spending by nearly $889 million to $1.32 billion, depending on the type of dispute settlement mechanism adopted by lawmakers. This translates into an additional cost of between $84.75 and $126.03 for each North Carolinian and a decrease in state gross domestic product.

North Carolina: Union Desert

According to the Bureau of Labor Statistics, North Carolina is one of the least unionized states in the nation. In 2019, an estimated 102,000 (or 2.3 percent) of nearly 4.4. million employed workers in North Carolina were members of a labor union or an employee association. Only South Carolina’s 2.2 percent rate was lower. Additionally, 150,000 North Carolinians (3.4 percent of the workforce) had jobs that were covered by a union or employee association contract. Again, only South Carolina had a lower rate (2.7 percent).1 (See Figures 1 and 2.)

Since 2000, North Carolina’s unionization rate generally has been on the decline. In 2000, the unionization rate was 4.0 percent, which would be the highest rate in the ensuing two decades. The rate dropped to a low of 1.9 percent in 2014, rose to 3.4 percent three years later, and dropped again to its current rate of 2.3 percent. The  percentage of workers covered by a union contract followed a similar trajectory. In the year 2000, 4.8 percent of workers had union coverage. By 2014, the  percentage had dropped to 3.2 percent, exceeded 4 percent over the next four years, and dropped to its current rate of 3.4 percent.2 (See Appendices A and B for long-term state and municipal membership and coverage trends.)

These figures can be misleading, however. North Carolina’s overall unionization rate has dropped, but these statistics do not differentiate between public- and private-sector unions. As this study will show, public-sector unions dominate North Carolina’s union landscape, and their expansion through the elimination of collective bargaining provisions would come at a high cost to the taxpayers of the state.

Early State and Federal Responses to Unionism

Labor law in the United States begins with the National Labor Relations Act of 1935 (a.k.a. the Wagner Act), which laid the foundations for private-sector unionization, collective bargaining, collective action, and employee rights. At the same time, it placed restrictions on the actions that employers may take in response to worker unionization activities. Moreover, it established the National Labor Relations Board, an independent federal entity tasked with mediating labor disputes. In a 5-4 decision, the United States Supreme Court in 1937 upheld the Wagner Act in National Labor Relations Board v. Jones & Laughlin Steel Corp.3

In North Carolina, unionization and labor unrest preceded passage of the Wagner Act. Labor unions began forming in North Carolina around the turn of the century. The American Federation of Labor began recruiting textile workers in North Carolina as early as 1898. The Raleigh Central Labor Union was founded in 1900. The North Carolina State Federation of Labor, a federation of dozens of trade unions, held its first convention in High Point in 1901. The 1919 United Textile Workers Union strike at the Highland Park plant in Charlotte was one of the first large-scale labor union protests in the state. Ten years later, the National Textile Workers Union (NTWU) organized the Gastonia strike at Loray Mill, a violent confrontation that attracted international attention.4

After the passage of the Wagner Act, labor unions continued to make inroads in North Carolina’s textile, furniture, and tobacco industries. The Textile Workers Union of America organized unions at 10 Fieldcrest mills in Rockingham County between 1939 and 1941 and at the Harriet and Henderson mills in Henderson County. In the early 1940s, Local 22 of the Food, Tobacco, Agricultural, and Allied Workers–Congress of Industrial Organizations organized thousands of tobacco manufacturing workers at R.J. Reynolds Tobacco Company in Winston-Salem. In May 1946, the Congress of Industrial Organizations launched “Operation Dixie” to advance the cause of organized labor in the South, an effort that concluded seven years later with minimal success.5

Following labor unrest throughout the United States in 1945 and 1946, Congress passed the Labor Management Relations Act of 1947 (a.k.a. the Taft–Hartley Act). The legislation was an effort to restrict certain activities by labor unions, including certain kinds of strikes, boycotts, picketing, and work stoppages. Unions were prohibited from donating directly to federal campaigns, and the law outlawed closed shops (the requirement that an employer hire union members only). President Harry Truman vetoed the bill two weeks after its passage, but Congress quickly and easily overrode the veto within days of receiving it.6

The Taft–Hartley Act authorized states to pass right-to-work laws, which prohibit union security agreements between employers and unions. The practical effects of the prohibition are significant, as it limits the ability of unions to require membership, dues, or conditions of employment on new and existing employees.

North Carolina was among a wave of states to pass a right-to-work law in anticipation of the Taft–Hartley Act. The North Carolina General Assembly passed House Bill 229 in March 1947, preceding passage of the Taft–Hartley Act by approximately 10 weeks. House Bill 229 prohibited contracts that required membership in labor unions as a condition of employment and banned unions from requiring membership or union dues for current and newly hired employees. The legislation also authorized employees to sue employers for damages if they were denied employment or fired from an employer that compelled employees to join a union.7 (See Appendix C.)

Congress imposed additional restrictions on labor unions with the passage of the Labor Management Reporting and Disclosure Act of 1959 (a.k.a. the Landrum–Griffin Act). The law mandated secret elections, reporting requirements, and additional worker protections. During the same year, Wisconsin became the first state to approve public-sector collective bargaining legislation.8

The North Carolina General Assembly approved a ban on collective bargaining by public-sector employees in 1959. The statute in Chapter 95, section 98 of the North Carolina General Statutes (G.S. § 95-98) reads:

Any agreement, or contract, between the governing authority of any city, town, county, or other municipality, or between any agency, unit, or instrumentality thereof, or between any agency, instrumentality, or institution of the State of North Carolina, and any labor union, trade union, or labor organization, as bargaining agent for any public employees of such city, town, county or other municipality, or agency or instrumentality of government, is hereby declared to be against the public policy of the state, illegal, unlawful, void and of no effect.9 (See Appendix C.)

This legislative effort to ban collective bargaining was led by Democrat Rep. Frank Snepp of Mecklenburg County. Snepp proposed House Bill 118 as a response to union organizing of Charlotte police by Jimmy Hoffa and the Teamsters. The city dissolved the union, and a lawsuit followed. In February 1959, Snepp declared, “There has been an obvious national trend to unions to move in on local government, especially police departments, and just as obviously it must be stopped.”10 Opponents of the bill included local, state, and national unions and a handful of North Carolina lawmakers who accused bill sponsors of doing the bidding of Charles Cannon of Cannon Mills in Kannapolis, a massive textile operation that had been (and would continue to be) a perennial target of union organizing efforts.

The Emergence of Public-Sector Unionism

Despite moderate success in organizing private-sector workers, few state and local employees joined labor unions during the first half of the 20th century. According to one estimate, less than 15 percent of the public-sector workforce was unionized before the 1960s, prompting a president of the AFL-CIO to remark once that it “is impossible to bargain collectively with the government.”11 Yet as pioneering labor economist Richard Freeman points out, by 1980, 43 percent of government employees had union representation, 32 percent had contractual agreements, 39 states permitted collective bargaining, and eight states authorized strikes.12

North Carolina was not one of them.

By 1984, North Carolina was one of two states (Virginia was the other) that expressly prohibited collective bargaining for state employees, police, firefighters, teachers, and other local employees.13 It was one of nine states that had no union security provisions, such as laws authorizing agency shops, unions shops, and dues checkoff. North Carolina was one of 25 states that had no mediation, fact-finding, or arbitration provision to resolve disputes between public-sector employers and unions. Finally, it was one of three states that had no explicit strike policy for any public-sector employee group. Unsurprisingly, North Carolina had one of the most inhospitable bargaining environments in the nation. Unlike Hawaii, Connecticut, Oregon, and other states that had strengthened their labor laws in the 1970s, North Carolina’s public-sector labor environment had changed little between 1969 and 1984.14

Even with these prohibitions on collective bargaining, public-sector unions in North Carolina hit their high-water mark by the mid-1980s, while private-sector unions struggled. The appetite for private-sector unionization in North Carolina was reflected in the refusal of textile workers at Cannon Mills to authorize union representation by the Amalgamated Clothing and Textile Workers Union in 1985 and futile attempts to unionize the mill in the 1990s. By 1984, over 20.3 percent of public-sector workers were members of a union, and a union agreement or contract covered over 26.5 percent.15 Among private-sector employees, only 4.8 percent were members, and an agreement or contract covered 5.6 percent.16 (See Figure 3.)

But these gains were not sustainable. By the mid-1980s, public-sector union membership and coverage had dropped significantly. In 1983, nearly 27.0 percent of public-sector union workers were covered by a union agreement. By 1986, the coverage rate had dropped to 17.6 percent. Coverage rates stabilized during the first few years of the 1990s, only to begin a sharp descent around the time of the dot-com bust. The 12.3 percent public-sector coverage rate in 1997 was the lowest in two decades.

Coverage rates remained low throughout the recession in the early 2000s, but rates soared during and after the Great Recession. Nationally, in 2009, the number of public-sector union employees exceeded their private-sector counterparts for the first time, although membership rates had favored public-sector unions for decades prior.17 With a few exceptions, North Carolina’s public-sector unions achieved numerical supremacy around the same time. By 2018, rates dropped again, diving to below 10 percent for the first time and fell to their lowest point going back to 1983. (See Figure 3.)

Union membership losses and the corresponding decline in political clout may have been among the factors driving the increase in public protests and related activities. On May 16, 2018, teacher union leaders achieved their most significant victory when North Carolina joined West Virginia, Kentucky, Oklahoma, Colorado, and Arizona as a “Red for Ed” walkout state. An estimated 30,000 public school employees and advocates traveled to Raleigh to participate in a school walkout sponsored by the North Carolina Association of Educators (NCAE), the state affiliate of the National Education Association and the largest teacher organization in the state. Despite years of eschewing the “union” label, the NCAE has embraced it. Days before the walkout, the NCAE Organize 2020 Caucus wrote, “This isn’t union-like activity. It’s union activity.”18

In 2019, the NCAE organized another teacher walkout, despite Republican leaders in the North Carolina General Assembly awarding teachers a sixth consecutive pay raise in the budget.19 The union marchers published a costly list of demands ahead of the walkout, including Medicaid expansion and a $15 minimum wage for all school personnel.20 The walkout took place on May 1, 2019, with significantly fewer participants than the 2018 event.

While the walkouts did little to produce membership gains,21 these activities may have contributed to increasing approval of labor unions in public opinion polls. Since the 1950s and ’60s, support for labor unions has been on the decline, although public sentiment appears to have changed in recent years. According to Gallup, around 75 percent of Americans supported labor unions in the 1950s, but support began to wane in the late 1970s. By 2009, support for labor unions fell to an all-time low of 48 percent, with an all-time high of 45 percent telling Gallup pollsters that they disapproved of labor unions.22

By 2019, labor unions had regained the public’s favor with a 64 percent approval rate vs. 32 percent disapproval rate.23 The effects of the COVID-19 pandemic on labor union support remain to be seen. Typically, support for unions fades during recessions, as both employed and unemployed workers question the costs and benefits of unions in the American economy.24 But the pandemic has a worker safety component that may bode well for unions that can convince employees that union membership would translate into robust health and safety protocols.

Amidst the resurgence of union activism and riding the wave of public favorability, efforts to repeal North Carolina’s collective bargaining statute were renewed by Democratic legislators in the North Carolina House and Senate. In 2019, Rep. Zack Hawkins (D-Durham) and Sen. Wiley Nickel (D-Wake) filed identical bills, House Bill 710 and Senate Bill 575, to repeal G.S. § 95-98.25 Those bills were two of the more recent efforts to repeal this statute, which have failed to reach the governor’s desk even in years with Democratic control of the legislature and governor’s mansion.26

Public-Sector Unions and Employee Associations in North Carolina

This study uses Barry T. Hirsch and David A. Macpherson’s “Union Membership and Coverage Database.” According to the database, North Carolina currently has an estimated 58,744 public-sector union members and 85,010 public-sector employees covered under a union contract.27 Self-reported membership counts solicited by the North Carolina Office of the State Auditor cannot be verified, are incomplete, and are only available since 2014. (See Tables 1 and 2.)

Despite these shortcomings, the State Auditor reports are useful for determining membership variations of the handful of public-sector unions and employee associations in North Carolina. In North Carolina, the State Employees Association of North Carolina (SEANC) is the largest employee association, with over 49,000 members. Over the last five years, membership in SEANC has dropped by a modest 7 percent and it is one of seven employee associations to post membership losses during this period. Four associations had membership gains, none larger than the 81.5 percent increase reported by the NC Public Service Workers Union. Total employee association membership dropped slightly over the last five years.

The second-largest employee association is likely the North Carolina Association of Educators (NCAE), historically one of the weakest teacher unions in the nation.28 The NCAE refuses to provide membership numbers to the State Auditor’s Office, which the agency says “does not have the authority to compel employees’ associations (private entities) to provide this type of information.” The Chapel Hill–Carrboro Federation of Teachers “never responded to requests for membership information,” according to the audit report.29

Presumably, the NCAE and Chapel Hill–Carrboro Federation of Teachers leadership refuse to cooperate because doing so would subject the organization to the provisions of Session Law 2014-115, which is the genesis for State Auditor data collection in the first place. That law allows employees of any local board of education to have dues withheld from their paychecks only if they are a member of a “domiciled employees’ association that has at least 40,000 members” (the majority of whom are public school teachers).30

According to the latest data published by its parent entity, the National Education Association, and data compiled by teacher union watchdog Mike Antonucci, the NCAE had 18,380 active members during the 2017-18 school year and 17,580 active members during the 2018-19 school year. (See Table 3.) Even if the NCAE included the 10,000 or so substitute, reserve, and retired staff memberships, the association would still have fewer than 30,000 total memberships and thus would forfeit the right to keep its 5,391 members on payroll deduction.31 Over the last five years, active membership has plummeted by 25 percent, despite more aggressive action and an embrace of the “union” label.

Literature Review

Since the 1960s, labor economists and researchers have focused on the wide-ranging political, economic, and social effects of public-sector unionization. In addition to tracking trends in union membership and coverage,32 studies have examined the relationship between unionization and political participation33 and worker proficiency.34 A more recent body of academic literature has focused on specific occupations and outcomes, such as the effects of public school unionization on such outcomes as student performance.35

While teachers are a population represented in this study, its primary focus will be on the public-sector unions generally. Moreover, while the author of this study offers descriptive statistics of union membership and coverage in North Carolina, it is not designed simply to track the trajectory of unionization in the state. (For a comprehensive digest of descriptive statistics, see Appendices A and B.) Rather, the author advances a model that estimates the cost of eliminating North Carolina’s collective bargaining law, which would expand the size and scope of government through higher wages for union members and those covered under union contracts. As mentioned in the previous section, efforts to eliminate the collective bargaining law are ongoing among some legislators in the North Carolina General Assembly. If a majority of lawmakers sympathetic to public-sector unions were elected, it could bring an end to a prohibition on collective bargaining that has remained in North Carolina statutes for 61 years so far.

The wage premium associated with unionization has been the subject of dozens of empirical studies since the 1980s.36 In a 1996 study published in Public Choice, Michael Marlow and William Orzechowski found that a more favorable legal environment for public unions increased government expenditures from $104.1 to $125.1 per capita to $546.0 to $653.5 per capita.37 In a more recent study, Marlow estimated that a 10 percentage point increase in public-sector union membership raised the size of government by 2.25 percent to 4.25 percent. Moreover, such an increase boosted real wages by an average of $1,603 (3.48 percent) per government worker.

More recently, in a 2011 Industrial and Labor Relations Review study of public school teachers, John Winters found that union activity increased salaries of experienced teachers by 18 percent to 28 percent, although beginning teachers enjoyed less of a boost.38 Also, Winters found a neighborhood effect whereby teacher salary increases in one district often lead to increases in nearby districts. According to his estimates, a 1 percent increase “in the distance-weighted average of teacher salaries in nearby districts increases salaries in a given district by 0.52 percent for experienced teachers and 0.66 percent for beginning teachers.”39

Not all studies conclude that public-sector unions enjoy a wage premium through the collective bargaining process. Brigham Frandsen’s 2016 causal effects study published in Industrial and Labor Relations Review concluded that collective bargaining laws have variable compensation and workplace effects on different occupations.40 Like most studies, Frandsen found that union membership increased when laws mandating bargaining existed. Nevertheless, he discovered that the effects on teachers, firefighters, and police were not uniform (forgive the pun). For example, public-sector collective bargaining rights appeared to have little impact on pay, benefits, or employment for public school teachers. Firefighters and police officers appeared to earn higher wages, although the primary benefit conferred on police officers was a shorter workweek. Frandsen concluded, “The results found in this article, however, suggest that the causal effects of collective bargaining rights may be more limited, especially for teachers.”

In a 2013 paper presented at the annual meeting of the American Economic Association, Richard Freeman and Eunice Han used Current Population Survey (CPS) data from the U.S. Census Bureau to compare teacher, police, firefighter, and state and local employee union membership and earnings across states with different collective bargaining environments.41 They included states that banned collective bargaining for public-sector employees along with states that mandated bargaining and allowed agency shop fees. Unsurprisingly, they affirmed that bans on collective bargaining led to lower membership rates than mandated bargaining. More importantly, they found that teachers’ wages increased and contract working days declined for union members in school districts, even for union members who live in states with laws that prohibit collective bargaining.

In sum, researchers generally agree that unionization increases wages and leads to marginal improvements in working conditions of public-sector workers, but the wage premium and other effects are not uniform across states or occupations. Regardless of these variations, increases in salaries, benefits, and employment necessitate higher aggregate government spending, although some union supporters hypothesize that unionization elevates job satisfaction and therefore mitigates turnover and the costs associated with it.42

Public-sector unions can increase government expenditures in two ways. The first is to use the collective bargaining process to negotiate higher salaries, more generous benefits, and expanded employment opportunities for members. But the effects are not just limited to the members of the bargaining group. Winters’ 2011 study summarized above suggested that the gains realized by bargaining parties appear to produce benefits that carry over to neighboring entities.

The second is to use the political process to support pro-union candidates for key local and state offices. Sympathetic elected officials can then use their regulatory and legal authority to protect union interests while using the power of taxation to raise revenues necessary to accommodate compensation and employment demands.43

Both public- and private-sector unions can use member resources to support sympathetic candidates. But public-sector unions depend on elected officials, rather than the market economy, for financial support. This dependency adds a powerful incentive for public-sector union leaders to immerse their organizations in political activity. According to one theory, the transformation of public-sector unions into organized political machines may have spurred the growth of public-sector unionization in the 1960s and 1970s.44

In a 2015 study published in The Journal of Politics, Sarah Anzia and Terry Moe examined increases in municipal government expenditures via bargaining and politics.45 In the first part of their study, spanning the 1970s and 1980s, the authors found that unionization of firefighters and police officers produced increases in average wages, employment, and payroll expenditures. In the second part of their study, spanning the 1990s and 2000s, Anzia and Moe concluded that collective bargaining led to higher spending on salaries and massive surges in health, hospital, disability, and life insurance benefits. Union activity in elections was associated with even larger expenditures.

This study will not attempt to differentiate public-sector unions’ gains through collective bargaining or the political process. I wish only to acknowledge that either leads to increases in government expenditures. Depending on the trajectory of North Carolina’s collective bargaining law, it may be a fruitful area for future research. Would a repeal of the collective bargaining law produce an acceleration of partisan political activity among public-sector unions, or does partisan political activity precipitate repeal?


This model was inspired by the work of Freeman and Han’s 2013 study “Public-Sector Unionism Without Collective Bargaining” cited previously.46 Freeman and Han found that the average union premium for a state is different depending on the occupation and the collective bargaining status of the state. Freeman and Han classified four different collective bargaining states: (1) the “High-CB [collective bargaining]” state, which mandates public-sector collective bargaining and “mandatory agency fees”; (2) the “Med-CB” state, which has compulsory public-sector collective bargaining but also has “right-to-work” provisions; (3) the “Low-CB” state, which allows for collective bargaining but does not require it; and (4) the “No-CB” state, which prohibits collective bargaining for public-sector workers. Freeman and Han also found that union density is different based on the public-sector occupation and a state’s collective bargaining status.47

According to the Annual Survey of Public Employment & Payroll (ASPEP) conducted by the U.S. Census Bureau for 2018, there were 558,973 full-time equivalent (FTE) employees in North Carolina. There were 172,461 FTE public school teachers, 26,043 FTE police officers, 9,067 FTE firefighters, and 351,402 FTE other state and local employees.48 Data from ASPEP 2018 and the conclusions from Freeman and Han were used to model the hourly wage for the changes in the collective bargaining status.49

As the state’s laws on collective bargaining change in the model, we control for two variables: the union hourly wages for each occupation and the unionization rate changes for each occupation. The Buckeye Institute model uses the average hourly wages found on two popular employment websites, Indeed and ZipRecruiter.50 This model only discusses wages and not additional benefits. This model also assumes the non-union hourly wage in North Carolina remains the same regardless of a collective bargaining law change and that there are no changes in work hours or employment. Future studies will look more particularly into how collective bargaining status also changes work hours and other forms of compensation.

As noted previously, North Carolina currently prohibits public-sector collective bargaining, unlike states with “duty to bargain” laws that allow or require some level of collective bargaining. This model assumes that a change from the “No-CB” state to the “High-CB” state would immediately change the union hourly wages as well as the unionization rate for each of the different occupations. These changes would be unlikely to occur immediately, but they would be possible in the long term.

If North Carolina were to move from a “No-CB” state to a “High-CB” state, The Buckeye Institute model finds it would be a significant cost to taxpayers. Hourly wages for teachers would surge by 9.1 percent, police officers’ wages would go up 9.0 percent, firefighters’ wages would increase by 7.5 percent, and all other state and local employees’ wages would go up 3.6 percent. Overall, public-sector hourly wages would increase by 5.3 percent. Moving from a “No-CB” state to either a “Low-CB” or a “Med-CB” state, however, would not appear to have a large effect overall on wages. While this model takes in many different assumptions, it does at least indicate a significant increase in hourly wages from moving to a “No-CB” state to a “High-CB” state. (See Table 4.)

The first step in determining the increase in public spending for North Carolina taxpayers as a result of “duty to bargain” laws is to determine the increase in union coverage as a result of particular collective bargaining policies.

There are two measures of public-sector unionism in a state. One is union membership, which is “the percentage of employed wage and salary workers who are unionized.”51 The other measure, union coverage, is “the percentage of employed wage and salary workers who are covered by union contracts.”52 Of the 694,333 FTE public-sector employees in North Carolina in 2019, 58,744 of them were union members, and union contracts covered 85,010 employees.53 This means the public-sector union membership rate for North Carolina is 8.5 percent. Also, 12.2 percent of North Carolina public-sector employees are covered by union contracts.54 In this study, we will be focusing on union coverage, not union membership.

In his 2005 study “Union Membership in the United States: The Divergence Between the Public and Private Sectors,” Henry S. Farber calculated the effect of laws defining collective bargaining rights on union coverage.55 Using his estimates, we calculate that, in comparison with a state that does not have any provision on collective bargaining, the probability that a public-sector worker would be covered by a union increases by 0.9 percent by being in a “collective bargaining prohibited” state.56

Looking at states that had a “duty to bargain,” the increase in public-sector union coverage depended on the “dispute settlement mechanism,” if there is one. Farber found there was a 4.8 percent to 6.7 percent increase in the probability of being covered by a union in comparison with a state that has no provision on collective bargaining.57 We can then subtract the “collective bargaining prohibited” state’s union coverage increase of 0.9 percent from those numbers. Therefore, a state that were to go from prohibiting collective bargaining to a “duty to bargain” state would increase its public-sector union coverage by 3.9 percent to 5.8 percent.

Given that North Carolina’s current public-sector union coverage rate is 12.2 percent (as of 201958), a “duty to bargain” would increase the union coverage rate to 16.1 percent to 17.9 percent.

Marlow and Orzechowski’s 1996 study examined the ways that public-sector union membership and coverage rates influence public spending. They found that for every additional percentage point in public-sector union coverage, public spending increased $9.13 per capita in 1985 dollars.59 Adjusting for inflation, the increase in spending would be $21.73 per percentage point per capita in 2019.

Using the data from Farber’s 2005 study, we can infer the increase in public spending as a result of a “duty to bargain.” We can multiply the inflation-adjusted public spending increases per percentage point per capita from Marlow and Orzechowski by the percentage increase in union coverage from Farber to get the increase in public spending per capita. Then multiplying that number by the population of North Carolina in 2019 (10,488,844 people60) would show the increase in public spending for North Carolina for 2019 if “duty to bargain” was in place.

Depending on the “dispute settlement mechanism” for the duty to bargain law, North Carolina public spending would increase between $889 million to $1.32 billion for 2019.61 (See Table 5.)

When calculating spending increases on North Carolina public-sector compensation from changes in collective bargaining policy, many factors need to be considered. We created a model to address some of these factors.

Tax Implications

Switching to a system of mandated public-sector collective bargaining would increase state-level government spending by at least $889 million. Paying for the increased cost of authorizing collective bargaining for public-sector workers would require various state tax increases.

Recent discussion in North Carolina has involved switching to a system of mandated public-sector collective bargaining, which would increase state-level government spending within a range of $889 million to $1.32 billion for 2019 and $918 million to $1.363 billion in 2022 dollars. Four state tax scenarios are provided for illustrative purposes only. They do not account for the effects of higher public-sector employee wages on county and municipal governments, which in practice may require lower increases in state personal income and corporate taxes and higher local property taxes and fees.

Scenario 1

The first scenario analyzes the effect of an increase of the statutory income tax rate under the adjusted baseline. As shown in Table 6, these changes increase rates so that combined static revenue estimates of the personal income increase are offset by changes in state-level government expenditures of the same magnitude.

Static estimates, however, do not incorporate economic responses (e.g., changes to business decisions and household behaviors) to the tax changes. The Buckeye Institute’s dynamic scoring model incorporates and accounts for economic responses and analyzes how tax policy changes would impact government revenues, economic activity, job creation, and business investment.

As a result of dynamic activity, tax revenue collected would be below the projected amount owing to a diminished economy (See Table 7, for example). Economic analysis using dynamic scoring is now used by the Congressional Budget Office and the Joint Committee on Taxation, who are the fiscal scorekeepers for the federal government.62

Before the Congressional Budget Office and the Joint Committee on Taxation incorporated full dynamic analysis, fiscal scorekeepers modeled responses to fiscal policy changes based on immediate impacts such as higher taxes on a good reducing the quantity demanded of that good. This traditional method of analysis did not, however, consider more widespread changes in macroeconomic activity that come about when the incentives of households and businesses are affected by new taxes. Full dynamic scoring analyzes how a policy change impacts the total economy, including important aggregate macroeconomic variables such as employment, consumption or investment.

The initial cost of the policy to implement required collective bargaining is estimated to be between $889 million and $1.32 billion for 2019. Under the traditional fiscal process, forecasters would estimate the tax increase needed to raise a required amount of revenue, accounting for small decreases in demand among consumers but not the wider economic effects that could also affect tax collections. For example, this kind of analysis could forecast a certain excise tax rate increase to cover a certain amount of increase in government spending, but it would make that forecast without fully considering how the change in the excise tax rate would affect buying behavior for other goods or the wider economy.

Because the dynamic model takes into account other economic variables, however, it shows that the projected tax revenue will fall short of the static revenue amounts, owing to lesser economic activity brought about by the personal income tax increase. As a result, this model finds that a shortfall would occur because the tax increase would raise less revenue than projected under normal fiscal forecasting methods. Table 7 presents the dynamic effects of Scenario 1 and reveals that these policy changes will lead to a $590 million decline in state gross domestic product (GDP) in 2022.

Scenario 2

The next part of the first scenario models the effects if revenue generation was split between the personal income tax and the corporate tax; specifically, if 75 percent of the revenue were drawn from the personal income tax and 25 percent from the corporate income tax. The personal income tax rate is raised to 5.51 percent, and the corporate income tax rate is raised from 2.50 percent to 3.16 percent. As shown in Table 8, the increase in static revenue would be deficit-neutral since all of the revenue raised is modeled as an increase in government spending.

Table 9 shows the effects of the taxes and spending changes according to the Buckeye Institute’s dynamic scoring model. These tax and spending policy changes will lead to a $780 million decrease in state GDP in 2022 and a $410 million drop in investment.

Scenario 3

Next, we model the upper bound of the state-level spending by an increase in the personal income tax rate from 5.25 percent to 5.74 percent. As shown in Table 10, the increase in static revenue would be deficit-neutral since all of the revenue raised is modeled as an increase in government spending.

Table 11 shows the effects of the taxes and spending changes according to the Buckeye Institute’s dynamic scoring model. These tax and spending policy changes would lead to a $880 million decrease in state GDP in 2022.

Scenario 4

Finally, the last part of the first scenario analyzes the combined effects of an increase in the personal income tax rate from 5.25 percent to 5.62 percent and an increase in the corporate income tax rate from 2.50 percent to 3.49 percent. Those increases would produce the largest impact on GDP and investment of all of the modeled scenarios. As shown in Table 12, the changes would increase rates so that combined static revenue estimates of the personal and corporate income tax increases would be offset by state-level government expenditures of the same magnitude.

The Buckeye Institute’s dynamic scoring model estimates, in Table 13, that these policy changes would lead to a $1.16 billion decline in state GDP in 2022.


For over 60 years, North Carolina has prohibited public-sector unions from collective bargaining. Any lawmakers mulling a repeal of the collective bargaining law should recognize that doing so would come at significant cost to the state.

The cost would depend on the type of collective bargaining law selected by lawmakers. The weakest of the four options, a duty to bargain with arbitration required, would incur the least cost (nearly $889 million). Next in cost would be two other types of collective bargaining systems — an implied but not explicit duty to bargain ($1.12 billion) and duty to bargain with strikes allowed ($1.21 billion). Finally, the costliest is the duty to bargain with fact-finding or mediation required ($1.32 billion).

Why do these differing approaches to dispute resolution produce differences in cost? As Freeman and Robert G. Valletta explained in 1988, “Laws favorable to collective bargaining produce higher wages by encouraging bargaining relations and by creating an environment in which covered and noncovered workers make wage gains.”63 In other words, certain dispute-resolution approaches are more advantageous for unions than others. For example, arbitrator decisions are binding on both parties, thereby mitigating the power of the union to use the process to realize gains in wages, hours, and terms and conditions of employment. Mediation and fact-finding are not binding on the parties. Although mediation and fact-finding requirements vary by state, the reports produced by fact-finders often are used by the unions to muster political support for the cause.

Note that these estimates model only state expenditures, not local expenditures, which would also be impacted by a repeal of the state’s collective bargaining law. For example, local tax revenues supply a significant portion of public school expenditures in North Carolina. Approximately 26 percent of operating expenditures for North Carolina public schools comes from local dollars.64

Presumably, the introduction of collective bargaining would lead to massive property tax increases at the local level with inconsistent returns on worker productivity and outcomes. Those costs and the county and municipal fiscal effects from an introduction of collective bargaining are beyond the scope of this study, but it would be a fruitful area for future research.

This study began with a series of questions about whether unionism would be a net positive or negative for North Carolina. In economic terms, a repeal of North Carolina’s ban on collective bargaining of public-sector employees would impose significant and recurring costs on taxpayers. Unions, employee associations, and their supporters argue that those costs would be offset with improvements in public services and economic benefits from increasing student achievement in public schools.

I am skeptical that embracing unionism in North Carolina would produce substantive improvements in those ways.

The primary goal of unions is self-preservation, often at the expense of the public purse.

Economist Thomas Sowell summarized this perspective succinctly: “The biggest myth about labor unions is that unions are for the workers. Unions are for unions, just as corporations are for corporations and politicians are for politicians.”65

Appendix A

Appendix B

Appendix C: North Carolina General Statutes Chapter 95: Department of Labor and Labor Regulations

Article 10.

Declaration of Policy as to Labor Organizations.

§ 95-78. Declaration of public policy.

The right to live includes the right to work. The exercise of the right to work must be protected and maintained free from undue restraints and coercion. It is hereby declared to be the public policy of North Carolina that the right of persons to work shall not be denied or abridged on account of membership or nonmembership in any labor union or labor organization or association. (1947, c. 328, s. 1.)

§ 95-79. Certain agreements declared illegal.

(a) Any agreement or combination between any employer and any labor union or labor organization whereby persons not members of such union or organization shall be denied the right to work for said employer, or whereby such membership is made a condition of employment or continuation of employment by such employer, or whereby any such union or organization acquires an employment monopoly in any enterprise, is hereby declared to be against the public policy and an illegal combination or conspiracy in restraint of trade or commerce in the State of North Carolina.

(b) Any provision that directly or indirectly conditions the purchase of agricultural products, the terms of an agreement for the purchase of agricultural products, or the terms of an agreement not to sue or settle litigation upon an agricultural producer’s status as a union or nonunion employer or entry into or refusal to enter into an agreement with a labor union or labor organization is invalid and unenforceable as against public policy in restraint of trade or commerce in the State of North Carolina. Further, notwithstanding G.S. 95-25.8, an agreement requiring an agricultural producer to transfer funds to a labor union or labor organization for the purpose of paying an employee’s membership fee or dues is invalid and unenforceable against public policy in restraint of trade or commerce in the State of North Carolina. For purposes of this subsection, the term “agricultural producer” means any producer engaged in any service or activity included within the provisions of section 3(f) of the Fair Labor Standards Act of 1938, 29 U.S.C. § 203, or section 3121(g) of the Internal Revenue Code of 1986, 26 U.S.C. § 3121. (1947, c. 328, s. 2; 2013-413, s. 15; 2014-115, s. 17; 2017-108, s. 20.5(a).)

§ 95-80. Membership in labor organization as condition of employment prohibited.

No person shall be required by an employer to become or remain a member of any labor union or labor organization as a condition of employment or continuation of employment by such employer. (1947, c. 328, s. 3.)

§ 95-81. Nonmembership as condition of employment prohibited.

No person shall be required by an employer to abstain or refrain from membership in any labor union or labor organization as a condition of employment or continuation of employment. (1947, c. 328, s. 4.)

§ 95-82. Payment of dues as condition of employment prohibited.

No employer shall require any person, as a condition of employment or continuation of employment, to pay any dues, fees, or other charges of any kind to any labor union or labor organization. (1947, c. 328, s. 5.)

§ 95-83. Recovery of damages by persons denied employment.

Any person who may be denied employment or be deprived of continuation of his employment in violation of G.S. 95-80, 95-81 and 95-82 or of one or more of such sections, shall be entitled to recover from such employer and from any other person, firm, corporation, or association acting in concert with him by appropriate action in the courts of this State such damages as he may have sustained by reason of such denial or deprivation of employment. (1947, c. 328, s. 6.)

§ 95-84. Application of Article.

The provisions of this Article shall not apply to any lawful contract in force on the effective date hereof but they shall apply in all respects to contracts entered into thereafter and to any renewal or extension of any existing contract. (1947, c. 328, s. 7.)

Article 12.

Units of Government and Labor Unions, Trade Unions, and Labor Organizations, and Public Employee Strikes.

§ 95-98. Contracts between units of government and labor unions, trade unions or labor organizations concerning public employees declared to be illegal.

Any agreement, or contract, between the governing authority of any city, town, county, or other municipality, or between any agency, unit, or instrumentality thereof, or between any agency, instrumentality, or institution of the State of North Carolina, and any labor union, trade union, or labor organization, as bargaining agent for any public employees of such city, town, county or other municipality, or agency or instrumentality of government, is hereby declared to be against the public policy of the State, illegal, unlawful, void and of no effect. (1959, c. 742.)

§ 95-98.1. Strikes by public employees prohibited.

Strikes by public employees are hereby declared illegal and against the public policy of this State.

No person holding a position either full-or part-time by appointment or employment with the State of North Carolina or in any county, city, town or other political subdivision of the State of North Carolina, or in any agency of any of them, shall willfully participate in a strike by public employees. (1981, c. 958, s. 1.)

§ 95-98.2. Strike defined.

The word “strike” as used herein shall mean a cessation or deliberate slowing down of work by a combination of persons as a means of enforcing compliance with a demand upon the employer, but shall not include protected activity under Article 16 of this Chapter: Provided, however, that nothing herein shall limit or impair the right of any public employee to express or communicate a complaint or opinion on any matter related to the conditions of public employment so long as the same is not designed to and does not interfere with the full, faithful, and proper performance of the duties of employment. (1981, c. 958, s. 1.)

§ 95-99. Penalty for violation of Article.

Any violation of the provisions of this Article is hereby declared to be a Class 1 misdemeanor. (1959, c. 742; 1993, c. 539, s. 667; 1994, Ex. Sess., c. 24, s. 14(c).)

§ 95-100. No provisions of Article 10 of Chapter 95 applicable to units of government or their employees.

The provisions of Article 10 of Chapter 95 of the General Statutes shall not apply to the State of North Carolina or any agency, institution, or instrumentality thereof or the employees of same nor shall the provisions of Article 10 of Chapter 95 of the General Statutes apply to any public employees or any employees of any town, city, county or other municipality or the agencies or instrumentalities thereof, nor shall said Article apply to employees of the State or any agencies, instrumentalities or institutions thereof or to any public employees whatsoever. (1959, c. 742.)

Article 13.

Payments to or for Benefit of Labor Organizations.

§ 95-101. Definition.

As used in this Article, the term “labor organization” means any organization of any kind, or any agency or employee representation committee or plan, in which employee or employees participate and which exists for the purpose in whole or in part, of dealing with employers concerning grievances, labor disputes, wages, rates of pay, hours of employment, or conditions of work. (1963, c. 244.)

§ 95-102. Certain payments to and agreements to pay labor organizations unlawful.

It shall be unlawful for any carrier or shipper of property or any association of such carriers or shippers to agree to pay, or to pay, to or for the benefit of a labor organization, directly or indirectly, any charge by reason of the placing upon, delivery to, or movement by rail, or by a railroad car, of a motor vehicle, trailer, or container which is also capable of being moved or propelled upon the highways and any such agreement shall be void and unenforceable. (1963, c. 244.)

§ 95-103. Acceptance of such payments unlawful.

It shall be unlawful for any labor organization to accept or receive from any carrier or shipper of property, or any association of such carriers or shippers, any payment described in G.S. 95-102 above. (1963, c. 244.)

§ 95-104. Penalty.

Any person, firm, corporation, association or partnership which or who agrees to pay, or does pay, or agrees to receive, or does receive, any payment described in this Article shall be guilty of a Class 3 misdemeanor and shall only be fined not less than one hundred dollars ($100.00), nor more than one thousand dollars ($1,000) for each offense. Each act of violation, and each day during which such an agreement remains in effect, shall constitute a separate offense. (1963, c. 244; 1993, c. 539, s. 668; 1994, Ex. Sess., c. 24, s. 14(c).)

Source: North Carolina General Assembly, North Carolina General Statutes (changes through September 27, 2019).

Appendix D: The Buckeye Institute Economic Research Center: Tax Model Methodology for North Carolina

For details, visit:


    1. Bureau of Labor Statistics, U.S. Department of Labor, “Table 5. Union affiliation of employed wage and salary workers by state,”
    2. Ibid.

    3. Justia, “NLRB v. Jones & Laughlin Steel Corp., 301 U.S. 1 (1937),”

    4. See Brent D. Glass and Wiley J. Williams, “Labor Unions,” NCPedia, See also Robert H. Zieger, ed., Organized Labor in the Twentieth-Century South (Knoxville: University of Tennessee Press, 1991). Brent D. Glass, The Textile Industry in North Carolina: A History (Division of Archives and History, North Carolina Department of Cultural Resources, 1992). Daniel J. Clark, Like Night and Day: Unionization in a Southern Mill Town (Chapel Hill: University of North Carolina Press, 1997). Jacquelyn Dowd Hall and others, Like a Family: The Making of a Southern Cotton Mill World (Chapel Hill: University of North Carolina Press, 2000). Leon Fink, The Maya of Morganton: Work and Community in the Nuevo New South (Chapel Hill: University of North Carolina Press, 2003). Robert R. Korstad, Civil Rights Unionism: Tobacco Workers and the Struggle for Democracy in the Mid-Twentieth-Century South (Civil Rights Unionism: Tobacco Workers and the Struggle for Democracy in the Mid-Twentieth-Century South, 2003). Travis Sutton Byrd, Unraveled: Labor Strife and Carolina Folk during the Marion Textile Strikes of 1929 (Knoxville: University of Tennessee Press, 2015). and Travis Sutton Byrd, Tangled: Organizing the Southern Textile Industry, 1930–1934 (Knoxville: University of Tennessee Press, 2018).

    5. Ibid.

    6. National Labor Relations Board, “Legislative History of the Labor Management Relations Act, 1947, Volume 1,” U.S. Government Printing Office, 1985, Google Books,

    7. See “Journal of the House of Representatives of the General Assembly of the State of North Carolina 1947” and “Journal of the Senate of the General Assembly of the State of North Carolina 1947,” North Carolina Digital Collections,

    8. Patrice M. Mareschal, “Public sector unions, democracy, and citizenship at work,” Labor History, 59, no. 1 (2018): 40, DOI: 10.1080/0023656X.2017.1375593.

    9. North Carolina General Assembly, “North Carolina General Statutes” (changes through September 27, 2019),

    10. Kays Gary, “Snepp Bill Would Bar City Unions,” Charlotte Observer, February 20, 1959, 1.

    11. Richard B. Freeman, “Unionism Comes to the Public Sector,” Journal of Economic Literature 24, no. 1 (March 1986): 45.

    12. Richard B. Freeman, “Unionism Comes to the Public Sector,” National Bureau of Economic Research, Working Paper No. 1452 (1984),

    13. Richard B. Freeman and Robert G. Valletta, “The NBER Pubic Sector Collective Bargaining Law Data Set,” Appendix B in When Public Sector Employees Unionize, Richard B. Freeman and Casey Ichniowski, eds., (Chicago: University of Chicago Press, 1988),

    14. Ibid.

    15. Barry T. Hirsch and David A. Macpherson, “Union Membership and Coverage Database from the Current Population Survey: Note,” Industrial and Labor Relations Review, Vol. 56, No. 2 (January 2003): 349–54 (updated annually at

    16. Ibid.

    17. Henry S. Farber, “Union Membership in the United States: The Divergence Between the Public and Private Sectors,” Working Paper #503, Princeton University Industrial Relations Section (September 2005): 1–29. Farber argues, “On balance, unions in the public sector have grown relative to unions in the private sector for important structural reasons. Lack of market competition for the products of the public sector and lack of fiscal discipline through the political process makes the value of unions to public sector workers relatively high.”

    18. Travis Fain, “Protest tactics leave bitter taste for some who favor more school funding,” WRAL (May 15, 2018),

    19. North Carolina Department of Public Instruction, Division of School Business, “Highlights of the North Carolina Public School Budget 2020,” (May 2020),

    20. T. Keung Hui, Martha Quillin, Dawn Baumgartner Vaughan, Anna Johnson, and Shelbi Polk, “For the second year, teachers march through Raleigh demanding more education funding,” News & Observer (May 1, 2019),

    21. Mike Antonucci, “Analysis: NEA Lost Almost 33,000 Working Members in 2019. Here’s the State-by-State Breakdown,” The 74 (June 17, 2020),

    22. Gallup, “Labor Unions,”

    23. Ibid.

    24. Dylan Scott, “The Catch-22 for labor unions enjoying newfound public support,” Vox, August 28, 2019,

    25. North Carolina General Assembly, “House Bill 710: Repeal Ban/G.S. 95-98,” 2019-2020 Session, North Carolina General Assembly, “Senate Bill 575: Restore Bargaining Rights,” 2019-2020 Session, See also Dawn Baumgartner Vaughan, “Freshmen state legislators want to overturn law that has held back unions for 60 years,” News & Observer, (April 24, 2019),

    26. Repeal bills were filed in 1989 (House Bill 868), 2007 (Senate DRS75348-LR-46A), 2009 (Senate Bill 427), and 2011 (Senate Bill 386).

    27. Hirsch and Macpherson, “Union Membership and Coverage Database.”

    28. A 2012 study of teacher union strength ranked North Carolina 40th out of 50 states and the District of Columbia. Amber M. Winkler, Janie Scull, and Dara Zeehandelaar, “How Strong Are U.S. Teacher Unions? A State-By-State Comparison,” Thomas B. Fordham Institute (2012),

    29. Office Of The State Auditor, “North Carolina Report Of Total Membership Count Of Employee Associations,” 2014 through 2019.

    30. North Carolina General Assembly, Session Law 2014-115,

    31. National Education Association, “Financial Reports Presented to the Representative Assembly” (July 2020).

    32. Casey Ichniowski, “Public Sector Union Growth And Bargaining Laws: A Proportional Hazards Approach With Time-Varying Treatments,” National Bureau of Economic Research Working Paper No. 1809 (1986). Barry T. Hirsch, David A. Macpherson, and John V. Winters. “Teacher Salaries, State Collective Bargaining Laws, And Union Coverage,” Association for Education Finance and Policy (AEFP) Meetings, Seattle, Vol. 26 (March 2011): 671–718.

    33. Patrick Flavin and Michael T. Hartney, “When Government Subsidizes Its Own: Collective Bargaining Laws as Agents of Political Mobilization,” American Journal of Political Science, 59, No. 4 (October 2015): 896–911. Heather Rosea and Jon Sonstelie, “School Board Politics, School District Size, And The Bargaining Power Of Teachers’ Unions,” Journal of Urban Economics 67 (2010): 438–450.

    34. Michael Lovenheim and Alexander Willén, “The Long-run Effects of Teacher Collective Bargaining,” National Bureau of Economic Research Working Paper No. 24782 (July 2018), Eunice S. Han, “The Myth of Unions’ Overprotection of Bad Teachers: Evidence from the District-Teacher Matched Panel Data on Teacher Turnover,” National Bureau of Economic Research Working Paper No. 83489 (January 2016),

    35. Benjamin A. Lindy, “The Impact Of Teacher Collective Bargaining Laws On Student Achievement: Evidence From A New Mexico Natural Experiment,” Yale Law Journal 120 (2010): 1130. Jason Baron, “The Effect of Teachers’ Unions on Student Achievement in the Short Run: Evidence from Wisconsin’s Act 10,” Economics of Education Review 67 (2018): 40–57. Joshua C. Hall, Donald J. Lacombe, and Joylynn Pruitt, “Collective Bargaining And School District Test Scores: Evidence From Ohio Bargaining Agreements,” Applied Economics Letters 24, no. 1 (2017): 35-38. J.M. Cowen and K.O. Strunk, “The Impact Of Teachers’ Unions On Educational Outcomes: What We Know And What We Need To Learn,” Economics of Education Review 48 (2015): 208–223. Eunice S. Han and Jeffrey Keefe, “Teachers’ Unions: Engaging Teachers and Improving Student Achievement in States That Prohibit Collective Bargaining,” Educational Policy (January 2020): 1–33. Morris M. Kleiner and Daniel L. Petree, “Unionism and Licensing of Public School Teachers: Impact on Wages and Educational Output,” in Richard B. Freeman and Casey Ichniowski, eds., When Public Sector Workers Unionize (Chicago: University of Chicago Press, 1988), 305–322. Michael F. Lovenheim, “The Effect of Teachers’ Unions on Education Production: Evidence from Union Election Certifications in Three Midwestern States,” Journal of Labor Economics 27, no. 4 (October 2009): 525–587. William H. Baugh and Joe A. Stone, “Teachers, Unions, and Wages in the 1970s: Unionism Now Pays,” Industrial and Labor Relations Review 35, no. 3 (April 1982): 368–376. John Thomas Delaney, “Strikes, Arbitration, and Teacher Salaries: A Behavioral Analysis,” Industrial and Labor Relations Review 36, no. 3 (April 1983): 431–446. Jordan D. Matsudaira and Richard W. Patterson, “Teachers’ Unions And School Performance: Evidence From California Charter Schools,” Economics of Education Review 61 (2017): 35–50. Todd E. Vachon and Josef (Kuo-Hsun) Ma, “Bargaining for Success: Examining the Relationship Between Teacher Unions and Student Achievement,” Sociological Forum 30, no. 2 (June 2015): 391–414. Kristine Lamm Westa and Elton Mykerezi, “Teachers’ Unions And Compensation: The Impact Of Collective Bargaining On Salary Schedules And Performance Pay Schemes,” Economics of Education Review 30 (2011) 99–108.

    36. See, for example, Gregg H. Lewis, Union Relative Wage Effects (Chicago: University of Chicago Press, 1986). Peter D. Linneman, Michael L. Wachter, and William H. Carter. “Evaluating the Evidence on Union Employment and Wages,” Industrial and Labor Relations Review 44 (October 1990): 34–53. Gregg Lewis, “Union/nonunion Wage Gaps in the Public Sector,” Journal of Labor Economics 8 (1990), s260–s328. David Blanchflower and Alex Bryson, “Changes over Time in Union Relative Wage Effects in the UK and the US Revisited,” National Bureau of Economic Research Working Paper No. 9395 (2002). Linneman, Peter D., Michael L. Wachter, and William H. Carter, “Evaluating the Evidence on Union Employment and Wages,” Industrial and Labor Relations Review 44 (October 1990): 34–53.

    37. Michael L. Marlow and William Orzechowski, “Public Sector Unions and Public Spending,” Public Choice 89, no. 1/2 (October 1996): 1–16.

    38. John V. Winters, “Teacher Salaries And Teacher Unions: A Spatial Econometric Approach,” Industrial and Labor Relations Review 64, no. 4 (2011): 747–764.

    39. Ibid., 762.

    40. Brigham R. Frandsen, “The Effects Of Collective Bargaining Rights On Public Employee Compensation: Evidence From Teachers, Firefighters, And Police,” Industrial and Labor Relations Review 69, no. 1 (January 2016): 84–112.

    41. Richard B. Freeman and Eunice S. Han, “Public sector unionism without collective bargaining,” Presented at the AEA Annual Meeting, San Diego, CA, January 6, 2013,

    42. See, for example, David Zonderman, “A last vestige of the Jim Crow era,” NC Policy Watch, April 29, 2009,

    43. Richard B. Freeman, “Unionism Comes to the Public Sector.”

    44. Joseph D. Reid and Michael M. Kurth, “The determinants of public sector bargaining legislation: comment,” Public Choice 66 (1990): 177–182.

    45. Sarah F. Anzia and Terry M. Moe, “Public Sector Unions and the Costs of Government,” The Journal of Politics 77, no. 1 (January 2015): 114–127.
    46. Richard B. Freeman, Richard Barry, and Eunice S. Han. “Public Sector Unionism Without Collective Bargaining,” 17 (Table 1).

    47. Ibid., 18 (Table 2).

    48. U.S. Census Bureau, “U.S. Department of Commerce, “Annual Survey of Public Employment & Payroll” (2018),

    49. Richard B. Freeman, Richard Barry, and Eunice S. Han. “Public Sector Unionism Without Collective Bargaining.”

    50. See salary data at, viewable at, and, viewable at

    51. Marlow and Orzechowski, “Public Sector Unions and Public Spending.”

    52. Ibid.

    53. Hirsch and Macpherson, “Union Membership and Coverage Database from the Current Population Survey.”

    54. Ibid.

    55. Henry S. Farber, “Union Membership in the United States: The Divergence Between the Public and Private Sectors,” Working Papers 882, Princeton University, Department of Economics, Industrial Relations Section (September 2005): 23 (Table 6). See also Henry S. Farber, “Nonunion Wage Rates And The Threat Of Unionization,” Industrial and Labor Relations Review 58, no. 3 (February 2005): 335–352.

    56. Farber, “Union Membership in the United States: The Divergence between the Public and Private Sectors,” 23 (Table 6, Column 2).

    57. Ibid.

    58. Hirsch and Macpherson, “Union Membership and Coverage Database from the Current Population Survey.”

    59. Farber, “Union Membership in the United States: The Divergence between the Public and Private Sectors,” 20 (Table 3, Column 6).

    60. U.S. Census Bureau, “Annual Survey of Public Employment & Payroll.”

    61. Farber, “Union Membership in the United States: The Divergence between the Public and Private Sectors.”

    62. Congressional Budget Office, “How CBO Analyzes the Effects of Changes in Federal Fiscal Policies on the Economy” (November 10, 2014), The Joint Committee on Taxation, “Macroeconomic Analysis At The Joint Committee On Taxation And The Mechanics Of Its Implementation. Outline Presentation Of The Joint Committee Staff At The Brookings Institution Program: Dynamic Scoring: Now What?” (January 26, 2015),

    63. Richard B. Freeman and Robert G. Valletta, “The Effects of Public Sector Labor Laws on Labor Market Institutions and Outcomes,” in When Public Sector Employees Unionize, Richard B. Freeman and Casey Ichniowski, eds., (Chicago. University of Chicago Press, 1988): 81–106,

    64. Information Analysis, Division of School Business, North Carolina Department of Public Instruction, “Highlights of the North Carolina Public School Budget” (May 2020): 4.

    65. Thomas Sowell, “Union Myths,” National Review (March 8, 2011),