Will Labor Fare Better Under State Labor Relations Law? Richard B. Freeman LERA Meetings January 2006 Acknowledgements: I benefitted from discussions with Joel Rogers and on joint work that will appear as R. Freeman and J. Rogers, “The Promise of Progressive Federalism,”in Making the Politics of Poverty and Inequality, ed. Jacob Hacker, Suzanne Mettler, and Joseph Soss (New York, NY: Russell Sage Foundation, forthcoming) This paper gives an affirmative answer to the title question. I argue that labor would fare better if the US reduced federal preemption of private sector labor relations law and devolved the legal regulation and enforcement of the right of association and collective bargaining to the states. The claim that labor would do better under state law than federal law runs counter to the standard story in US labor history and the traditional views of pro-labor scholars. US history warns against trusting states to protect labor rights. The southern states made slavery legal until the Civil War and used state laws to suppress blacks for over a century after the 14th Amendment. Given the chance to enact legislation harmful to unions by the Taft-Hartley Act, 21 states decided in favor of right to work legislation that forbids firms and unions from including union security clauses in their contracts. The archetypical villain in union organizing is a racist Southern sheriff who abuses his legal powers to put down the Norma Raes of the world. By contrast, labor histories often credit the National Labor Relations Act for the growth of unionism in the US in the late 1930s through the 1950s. 1 Congress’s enactment of Title IV of the Civil Rights Act of 1964 dealt a major blow to discrimination against blacks, particularly in southern states that would never have passed anti-discriminatory legislation on their own. From the 1960s through 2000, Congress enacted a host of other labor bills -- protecting different groups against discrimination, regulating occupational health and safety, insuring employment retirement plans, etc, so that as of 2005, the Department of Labor administered and enforced more than 180 federal laws covering 125 million workers.2 In the area of labor relations law, on which this paper focuses, courts have held that the federal legislation preempts state activity. States are free to legislate wages above the federal minimum and enact and enforce employment law, which covers individuals rather than groups, but are precluded from taking independent action on private sector labor relations issues. Much of the impetus for federal preemption came from liberal Supreme Court justices, influenced by pro-labor legal scholars such as Archibald Cox, who argued for federal preemption although there is no explicit Congressional statement in this regard.3 Indeed, since the Supreme Court decision in Garmon (1959) courts have held that the NLRB has exclusive jurisdiction over conduct regulated by the LRMA. More recent legal decisions have extended the preemption doctrine to efforts by states to limit the anti-union activity of 1 Even after passage of the Wagner Act, however, many workers unionized without using the Act. Freeman (1998). http://www.dol.gov/opa/aboutdol/lawsprog.htm 3 Michael H. Gottesman, “Rethinking Labor Law Preemption” Yale Journal on Regulation 7 (Summer 1990): 355410 2 2 their contractors. At the same time, the Supreme Court has interpreted state sovereign immunity to bar suits for damages by state employees for violations of federal employment rights.4 In recent years, however, labor has fared poorly under national regulations. From the 1960s through the 2000s, unions gained miniscule numbers of members through the NLRA election process. The Pension Benefit and Guaranty Corporation which insures private defined benefit pensions has run large deficits as the private system lurches from crisis to crisis. Relative to average hourly earnings, the federal minimum wage has fallen to its lowest level in decades. It is barely half the minimum wages in the UK and in Ireland, despite the US having 40 percent higher GDP per capita than those countries. My case that labor would do better under state law begins with the standard federalism arguments for devolving authority: that it allows experimentation in policies and creates space for divergent policies to match citizen preferences. I show that state law has produced wider cross-state variation in the unionism of public sector workers than federal law has produced in unionism of private sector workers. Associated with this is a higher level of unionisation in the public sector than in the private sector. I use evidence on the variation in state policies toward labor broadly to suggest that if states had the right to enact or enforce labor relations, some states would enact policies more favorable to collective bargaining than the Congress, while others would enact less favorable policies. Because private sector collective bargaining and unionisation is exceptionally low in the states likely to introduce legislation unfavorable to collective bargaining, such legislation cannot have much effect on union density or coverage. By contrast, a more friendly legal environment in the states that look favorably on collective bargaining has the potential for giving workers in those states the representation and participation at the workplace that they say they want. Finally, I argue that state legal enactment offer the best opportunity for the country to experiment with new forms of worker organizations, including variants of the company sponsored workplace organizations that the LMRA outlaws, to open the market for representation and participation to meet the desire of workers who want and cannot get unions under natiional law and of workers some nonunion organization short of unions to represent them to employers.5 Landau, Brent W. “State Employees and Soveriegn Immunity: Alternatives and Strategies for Enforcing Federal Employment Laws” Harvard Journal on Legislation, vol 39, pp 169-212 5 Every survey shows that millions of nonunion workers want unions to represent them but are unable to gain that representation. The 1994-95 Workplace Representation and Participation Survey and recent Hart Research Associates surveys show that even larger numbers of non-union workers want employee organizations that fall short 4 3 If my analysis is correct, the union movement would do better to spend its resources to press the Congress and courts to reduce or eliminate federal preemption of private sector labor law and to improve labor regulation and enforcement at the state and local level than to push for “labor law reform” at the national level. 1. The Logic for Devolving Labor Relations Law to states “ It is one of the happy incidents of the federal system that a single courageous state may, if its citizens choose, serve as a laboratory; and try novel social and economic experiments without risk to the rest of the country”. Justice Brandeis6 The classic argument for federalism has three parts. The first, as given by Brandeis above, is that federalism allows for experimentation so that the country can better assess the impact of different policies and then choose the one that works best in accomplishing the public purpose. States are well-suited to serve as laboratories of experiment because they are numerous, have boundaries that cover populations with similar attributes and views, and have the large scale necessary for assessing major policy initiatives. As a practical matter, policy changes are easier to achieve in a state than across the whole country, and reduce the risks inherent in any new initiative turning sour from the country to the state. The natural experiment where state A chooses policy I and neighboring state B chooses policy II can provide information on the aggregate effect of the policies which elude smaller random assignment studies. The second part of the argument for federalism is that states will imitate the most successful policy initiatives, so that good practices spread rapidly. The competitive market moves toward the right output in part by the exit and entry of consumers or firms. If, by chance, a given firm produces a product that consumers prefer to others, consumers will switch to it, and other firms imitate the lucky producer or go out of business. This is natural selection at work in the economy. Since states that choose the wrong policies do not go out of business, the market for policies operates somewhat differently. Citizens have the choice of moving to a jurisdiction with more desireable policies, or of taking political action to induce their jurisdiction to select their desired policies. Whether these forces operate effectively in any reasonable time period is unclear. The long association of states rights with slavery and disenfranchisement of blacks has made many Americans uneasy about trusting state of a collective bargaining union. See Freeman and Rogers, What Workers Want, Cornell, 2 nd edition, forthcoming, 2006, chapter 1). 6 Brandeis, dissenting, New State Ice Co. v. Liebmann, 285 U.S. 262, at 311 (1932). Greve critiques the case to which Brandeis dissented www.aei.org/publications/pubID.12743/pub_detail.asp. 4 power in the political area. Conservatives often express unease about state fiscal responsibility – will states will run up huge deficits that require a national bailout or possibly default? (Ferejon and Weingast) Liberals worry that the potential mobility of capital will lead states into a race to the bottom, as each state reduces its protections for labor to attract investments. After all, don’t states compete for particular investments through tax holidays and subsidies that benefit the investor more than citizens? (Donahue) But the politics could go the other way. Citizens in a state with worse labor relations policies might see how well people are doing in a state with better labor regulations, and move to the more desireable state or campaign for their state to copy the preferable legislation. Federalism in welfare services and medical services for the poor has not led to any race to the bottom among the states. Because states deliver services to citizens subject to hard budget constraints, governors and legislators tend to be less ideological and divisive than representatives and senators in Washington. State governments choose policies that make pragmatic sense even when those policies run against the prevailing ideology in their party. In 2004, the Republican legislature in Nebraska, for instance, chose a defined benefit state pension system over a defined contribution system because of evidence that this would save taxpayers money, while the Bush Administration was pushing for private accounts to replace Social Security. Historically, many of the basic social insurance, worker rights, public health, administrative and government reforms adopted by the New Deal were first anticipated and developed in states, such as Wisconsin. These policies spread to other states, and then became national law. The third part of the argument for federalism is that in a large diverse country, federalism allows each area to find the policies that fit it best. Mississippi gets the policies that it wants. Minnesota gets the policies it wants. California does whatever California wants. New York is New York. And so on. Forcing every state to follow the same rules and regulations necessarily reduces well being. In standard economic models, moreover, high geographic mobility increases the gains from policy diversity, since people with preferences that differ from those in their locale can move to a different locale. This is the Tiebout solution to the optimal determination of state and local expenditures. Citizens opposed to the labor regulations in their state vote with their feet by moving to states whose policies satisfy their preferences. Geographic mobility is the key market force, comparable to consumer choice in a competitive market. 5 The conceptual case for federalism is a strong one, which has attracted increasing academic support in the past decade.7 Still, there are potential benefits to national legislation – incorporation of externalities across state borders, maintenance of a single market, economies of scale in setting policy -- that argue in the other direction. In addition, federal law can spread best practice policies to states that might lack the political will or funds to protect or help their own working citizens. Given that there are plausible arguments on both sides of the debate, which may weigh more heavily in some situations than in others, the question of whether labor will do better under state law cannot be answered in the abstract. It requires evidence on how states have actually regulated labor relations when they have authority over labor laws or policies; and of the impact of their actions on the ability of workers to gain the participation and representation at their workplace that they want. 2. Variation in State Labor Relations Law In one respect, the US labor relations system is well-suited for examining the effect of state as opposed to federal regulation of labor relations. This is because states have authority over labor relations for state and municipal workers. Federal legislation sets and enforces the rules for private sector workers but state legislation sets and enforces the relations laws for public sector workers. This means that every state has private sector workers covered by federal law and public sector workers covered by state law, possibly in the same occupations or activities. Since state law can vary greatly, this provides a “natural experiment” in which we can contrast outcomes for the treatment group, public sector employees covered by that state’s public sector laws, against the control group of a private sector employees in that state covered by the ubiquitous federal law. To see how states regulate the state and local public sector workers over whom they have jurisdiction, I have examined the Kim Rueben update of the NBER Valletta-Freeman state public sector labor law data set (http://www.nber.org/publaw/). This data set categorizes public sector labor laws from 1955 to 1985 for five groups of workers (state employees, local police, local firefighters, local teachers, other local employees) in tems of their legal treatment of collective bargaining and the right to strike. Rueben has updated the data set through 1996. Valletta and Freeman found huge variation in public sector labor laws among states and smaller variation among groups of workers within states. They showed that the legal environment changed greatly in the 1960s and 1970s 7 See the Boston Federal Reserve New England Economic Review, June/July 1998 http://www.bos.frb.org/economic/neer/neer1998/neer398b.pdf 6 when many states shifted from meet and confer laws to laws that either explicitly or implicitly required government employers to bargain with unions. Table 1 summarizes the legal status of public sector collective bargaining in the 50 states in 1996, the last year of the Rueben update.8 The table places the laws governing state/ occupation groups into 3 broad categories: having a favorable legal environment for public sector collective bargaining – where laws explicitly require or imply that public sector employers have a duty to bargain with unions; an intermediate environment where the law requires public sector employers to meet and confer with unions but not to bargain with them; and an unfavorable legal environment, where the law either explicitly outlaws bargaining or contains no provision for bargaining. According to the table, sixty-four percent of the state occupation groups have laws favorable to collective bargaining. The bulk of these observations are found in the twenty-seven pro collective bargaining states listed in the table. Fourteen percent of the observations occur in state occupation groups covered by meet and confer laws, largely in the five states listed in the table. Finally, 22% of the observations have laws that are unfavorable to collective bargaining. These are concentrated in 17 states, many in the South. Since 1996 the legal changes governing state employees have been modest, though there are notable developments in particular states. 9 The states whose legal stance is unfavorable to public sector collective bargaining overlap to a considerable degree with the states that have outlawed union security clauses in private sector collective bargaining. Twenty one states have enacted such right to work (RTW) laws. 10 Fifteen states with RTW have public sector labor regulations in the “least favorable to collective bargaining” category. Two RTW states have regulations that fit into the intermediate “meet and confer” grouping. Four have regulations that are favorable to collective bargaining. 11 In percentage terms, 71% of the right to work states are in the least favorable category for public sector collective 8 For the most recent data on state regulation of teachers see Krueger, Carl, State Collective Bargaining Policies for Teachers http://www.ecs.org/clearinghouse/37/48/3748.pdf 9 Clark, Jr. R. Theodore PUBLIC SECTOR COLLECTIVE BARGAINING: PAST, PRESENT AND FUTURE,http://www.npelra.org/articles/collective_bargaining.asp 10 Alabama, Arizona, Arkansas, Florida, Georgia, Idaho, Iowa, Kansas, Louisiana, Mississippi, Nevada, Nebraska, North Carolina, North Dakato, Oklahoma, South Carolina South Dakota, Tennessee, Texas, Utah, Virginia 11 Here is the cross tabulation of RTW laws by the legal environment for public sector bargaining: Public Sector Legal Environment toward collective bargaining Unfavorable Intermediate Unfavorable 0 1 2 RTW 15 2 4 No RTW 6 3 23 with the definition of public sector legal environment as in table 1. 7 bargaining. On the basis of their actions, I would expect that if the states that have legislation unfavorable to public sector collective bargaining and RTW laws were to regulate private sector labor relations, they would create an unfavorable legal environment for unions and collective bargaining in the private sector as they have done in the public sector. On the other side of the divide, 72% of non RTW states are in the most favorable category for public sector collective bargaining, which suggests that they might institute more favorable regimes for collective bargaining than the national government has done. How have the different public sector labor regimes affected the prevalence of collective bargaining in a given state? To see the extent to which differences in public sector labor are related to public sector unionization and collective bargaining, I have linked the public sector labor law measures to estimates of the percentage of public sector workers covered by collective bargaining from the CPS, as given by Unionstats.com (http://www.unionstats.com). Panel A of Table 2 compares public sector bargaining coverage and private sector bargaining coverage among states with different public sector bargaining laws. The figures in the line public sector, mean coverage, show that states with favorable laws to collective bargaining have more than twice the coverage as states with unfavorable laws, while those with intermediate laws have coverage rates just modestly higher than those with unfavorable laws. The figures in the line private sector, mean coverage, show much smaller differences in density for the private sector. The figures in the line within-state differences give the difference between the public sector density and private sector density by state. States with favorable public sector collective bargaining laws have statistically significant higher differences between public and private density. The regressions in panel B of table 2 exploit the within state variation in unionization and laws to examine whether state laws more favorable to collective bargaining are associated with greater collective bargaining coverage in 2004, conditional on private sector unionism in the state and the presence of a Right toWork Law, using a linear and a log-linear regression model. Line 1 gives the univariate regression of public sector coverage on dummy variables for whether the state had public sector laws that were either favorable or unfavorable to collective bargaining; the deleted group are the states with intermediate laws. It replicates the mean differences in panel A in showing the huge impact of favorable laws on collective bargaining coverage in the public sector. Line 2 adds the 8 level of private sector bargaining and whether or not the state has a right to work law as additional independent variables. If public sector bargaining was due to the state’s general attitude toward unionism, as reflected in the private sector unionization rate and decision to enact an RTW law, the public sector favorable law variable would lose its significance. It does not. The level of private sector coverage has a sizeable significant effect on the percent covered, while the presence of a right to work law has a moderate negative effect. Lines 3 and 4 transform the coverage variables into ln form in order to get an estimate of elasticity of coverage to having favorable or unfavorable labor laws. The coefficient on the favorable law variable is the 0.50 in line 4. This suggests that moving from an intermediate legal status to a favorable status for collective bargaining raises coverage by 0.50 ln points or about 65% (= exp 0.50). To be sure, this cross section relation does not establish causality. If Missisisippi enacted a public sector labor law comparable to Wisconsin’s, its public sector might not have the same level of collective bargaining as Wisconsin. But the cross section creates a presumption of some effect. And studies of public sector unionization before and after changes in state labor laws in the 1970s-1980s suggest that the laws had an independent effect on unionization, or at least on the timing of union growth in the public sector (see Farber, 1988, Saltzman, 1988,Valletta and Freeman, 1988). Farber 2005 uses the within-state variation in laws by type of worker and finds that union density is signicantly higher where unions are allowed to negotiate union security provisions (e.g., agency shop) and where employers have a legal duty to bargain with labor unions. Finally, under the standard argument for federalism, the variation in state regulations of public sector bargaining ought to produce greater dispersion in the percentage of state and local employees covered by collective bargaining than the same federal law would produce in the percentage of private sector employees covered by collective bargaining. In 2004 the standard deviation of public sector union coverage rate across states was 16.5 whereas the standard deviation of private sector union coverage was 3.8. But since the public sector rate of unionization was much higher, its standard deviation was almost certainly higher as well. A tougher test of the claim that variation in state regulations produces the greater variation in coverage in the public sector than in the private sector is to compare the standard deviation of each sector with the standard deviation from a binomial distribution, with the mean rate of coverage across states taken as the probability of coverage. For the public sector the actual standard deviation is 2.44 times the expected standard deviation. while for the private sector, the actual 9 standard deviation is 1.00 times the expected standard deviation. Thus, the public sector shows statistically significantly more variation than one would expect from the binomial model while the private sector shows the same variation one would expect from the binomial model. The only plausible explanation is the difference in legal labor law regimes. To be sure, there are reasons for public and private sector union density to diverge even under the same legal regime. The profit motive ought to induce greater opposition to unions among private sector employers than among public sector employers, for whom union members and their allies can be an important part of the electorate. Workers may have different desires for unions in the two sectors, depending on lengths of employment, civil service regulations, and market or political pressures. Still, the difference in the legal regimes has been associated with trends in collective bargaining coverage that have produced huge differences in the level of bargaining between workers covered by state regulations and those covered by federal regulations. As of 2005, private sector union density has fallen to 7-8% of the work force, while public sector density has stabilized at about 35% of the public sector work force (see Farber, 2005 for a detailed analysis of the changing fortunes of unions in the two sectors). As table 2 shows, the gap between the public sector and the public sector is greatest in states with favorable public sector labor laws. Since private sector unionism declined in virtually all states at roughly similar rates in the 1980s and 1990s, while public sector density has stabilized, the gap between the public/private sector increased most in the states with favorable public sector labor laws. Finally, the experience of Canada with provincial determination of much labor legislation is also consistent with the view that organized labor does better when labor laws are set at lower levels. Union density is falling in Canada, but much more gradually than in the US. At various times provinces have enacted card check legislation that US unions would regard as dream “pie in the sky” to get from the US Congress. Others have enacted less favorable legislation. There are sufficient changes in legislation in the same province over time to show that the provincial legislation affects unionisation rates. 3. Variation in Other Labor Regulations That organized labor does better in the public sector with state laws governing state and local workers than it does in the private sector with national laws governing private sector workers is consistent with the claim that labor would do better under state regulations. But it does not establish the claim since, as noted, there are other factors that differentiate public and private sector employer and employee behavior. To assess the plausibility of the 10 claim further, I examine next state regulations of private sector labor workers in areas where states have authority. If there is considerable variation in state laws governing private sector labor and if those laws are more favorable to labor in states that have public sector regulations favorable to collective bargaining, it would seem reasonable to expect that those states would create a legal environment favorable to private sector unionism if they had authority over the private sector. By the same logic, the opposite would be true for states that had unfavorable public sector regulations and enacted laws less favorable to private sector labor in areas of state jurisdiction. Devolving private sector labor relations law to the states would thus produce legal environments that would make private sector collective bargaining easier in some states and more difficult in others, so that the key question becomes what happens on average among the states. Table 3 gives the evidence that there is considerable variation in state laws governing private sector labor and that the laws are more favorable to labor in the states that have also enacted favorable public sector collective bargaining laws. The table focuses on eight laws covering different aspects of labor regulation in the private sector, ranging from minimum wages to transitional food stamps for families leaving welfare. 12 In each case, having the regulation means that the state is more favorable to workers. To keep the statistics consistent, I coded the two measures that are naturally numeric – the criterion for medicaid eligibility, and support for need based aid for postsecondary students – as dichotomous variables around the cut point specified . There are other measures of state legal and administrative activity that impact workers, such as the existence and operation of state equal opportunity offices, the activities of state departments of labor, workmen’s compensation, and so on, that might be added to the list. Column 1 of the table records the number of states which have adopted each of the particular policies, while columns 2-3 show the number of states with the policy in states that have favorable, and intermediate/unfavorable public sector bargaining laws. For ease of presentation I have collapsed the intermediate and unfavorable states together. The percentages in each box measure the percentage of states in that category that had the particular law in the row. For instance, the line minimum wages above federal level shows that 16 states had above federal minimum wages in 2005 and that all 16, or 100%, were in states with favorable public sector labor laws. Since there were 27 states with favorable public sector labor laws, 59% of them had the law whereas 0% of 12 The eight laws exceed the number given by PERI in their analysis of how states treat labor or the Fraser Institute indicator of labor market freedom,. Both of these indicators mix laws with outcomes from other policies. 11 the states in the intermediate/unfavorable group had minimum wages above the federal level. The next line 2 shows that 15 states offered medicaid to workers with earnings over 100% of the poverty line and that 13 were in the states with favorable public sector labor laws while 2 were in intermediate/unfavorable group of states. This gives probabilities of having that law of 48% and 9% By way of summary, at the bottom of the table I give the total number of the laws in each group of states and the average number of laws adopted in each group. The states with favorable public sector collective bargaining laws averaged 4.3 laws compared to 2.2 laws for the other states. The differences between the favorable and intermediate and unfavorable states in private sector labor regulastions is statistically significant. Finally, looking at the entire distribution of pro labor laws/policies by state, I counted the number of prolabor laws/policies that each state had and compared that count to the distribution that would have resulted if the states had drawn their policies/laws from an urn in which the probability of a pro-labor policy was the probability of having a law, 45% (= 179/400, the total number of possible adoptions of a labor-favorable law). The actual distribution differed greatly from what one would expect if states randomly, with much greater numbers of observations in the tails. For instance, Wisconsin and California had most of the laws/ policies, while several had one or none of the policies. In sum, states policy stances toward private sector labor vary considerably in ways associated with the way they regulate public sector collective bargaining. Some states are favorable to labor interests and likely to enact or enforce laws favorable to labor in the private sector as they do in the private sector labor domains over which they have authority. On the basis of their legal stance in the public sector and pro labor legislationi in the private sector, I would expect them to create favorable legal environments for collective bargaining in the private sector. Other states are less favorable to labor interests in public sector collective barganinng law and in laws or policies for labor in the private sector. They are likely to create a more hostile legal environment for collective bargaining than under federal legislation. How these two forces would work out for the country as a whole depends on the nature of the shift from federal to state law and the numbers of workers in the groups of states. 4. The Net Outcome There is one way in which devolving labor law from the federal government to to state government would guarantee that labor gains. This would be if the devolution follow edthe Taft-Hartley model of allowing states to go beyond federal rules in one direction only. Just as the Taft-Harley law allows states to outlaw union security 12 provisions but does not give them any way to enact laws to help protect workers rights of association or to strengthen collective bargaining, Congress could give states the right to pass laws or enforce existing law onlyin ways that enhanced worker rightscollective bargaining. Some states would surely act on this. In the 1980s, Wisconsin directed its Department of Industry, Labor and Human Relations to list firms who violated the LMRA over a five year period and to bar those firms from state procurement. In 1986 the Supreme Court unanimously ruled that by prohibiting state purchases from repeat labor law violators, the state was acting as a regulator rather than as a purchaser and held that the LMRA preempted the Wisconsin statute.13 In 2000 California tried to get around this rule by enacting a law that prohibited employers from using funds received by the state to oppose union organizing. 14 The emphasis was on the state’s rights as a purchaser rather than as an entity seeking to enforce the LMRA . In 2002 New York enacted similar legislation, and in 2003 the Hawaii AFL-CIO Federation proposed a similar statute for Hawaii. Employers challenged the California law on the notion that it limited the legal rights of employers under the LMRA to speak against unionization. The ways courts have interpreted federal preemption it is dubious that any of these efforts will succeed. In 2002 the US District Court for California ruled against the state, interpreting employer free speech to include the right to use state contract moneys to fight unionization even when state legislatures explicitly passed laws that said they do not want taxpayer money spent in that way. Absent a Supreme Court decision reversing this and other preemption decisions15, Congressional action would be needed to create space for states to operate in ways that they have demonstrated they would like to do But if Congress were to act, it would more likely allow states to undertake policies that could weaken as well as strengthen unionism. Such a weakening of federal preemption could arguably gain the support of representatives from states supportive of unionism and of representatives from states antithetical to unions. The most plausible outcome would thus be increased diversity in state laws regulating private sector unionization, 13 WISCONSIN DEPT. OF INDUSTRY v. GOULD INC., 475 U.S. 282 (1986) 475 U.S. 282 Argued December 9, 1985 Decided February 26, 1986 (www.usscplus.com/on.line/475/4750282.htm). 14 California Assembly Bill, AB1889, Cedillo, 00-872, State Funds: Unionization (2000). 15 The Supreme Court could respond to the California case by following the 1993 Rehnquist Court decision that allowed Massachussetts to exclude nonunion contractors in the Boston Harbor construction case Building & Construction Trades Council v. Associated Builders & Contractors of Massachusetts, 507 U.S. 218 (1993), known as “Boston Harbor”, which would reduce the need for Congressional action to weaken federal preemption. 13 mimicking to some extent the diversity in public sector labor relations. Some states might introduce card checks for union recognition. Others might try the quick elections that the Dunlop Commission recommended. Others might enact place greater administrative regulatory burdens on unions, or discriminate against unionized firms in its state contracts or increase the rights of employers to oppose unions. The answer to the title question hinges, then, on how many workers might gain stronger rights to association and how many might see their rights weakened. To answer the question, I compare estimates of the potential loss of coverage in states most likely to enact state laws unfavorable to private sector collective bargaining with the potential gain in states most likely to enact laws favorable to private sector unionism. My estimates are back-of-the envelope/excel computations that give orders of magnitudes only. For my estimates I assume that all states with unfavorable public sector labor laws enact equally unfavorable laws in the private sector (though they would presumably not be allowed to outlaw collective bargaining). Column 1 of table 4 shows that in 2004, unions gave collective bargaining to 1.2 million persons in these states. To see how many of these members might lose collective bargaining as a result of a change to a legal regime less friendly to collective bargaining, I apply the elasticity of 0.5 from the regression of collective bargaining in the public sector to favorable public sector laws given in table 2, line 4. My assumption is that a more unfavorable environment would reduce coverage in the private sector by the same ln amount as favorable public sector laws boosted coverage in the public sector. The calculation implies a loss of 0.5 million persons. With just 1.2 million members in these states, the maximum loss is 1.2 million. The second column of table 4 estimates the possible gain to unions from a more favorable private sector legal environment in the states that have favorable public sector labor laws. Again, I use the elasticity of coverage of 0.5 from table 2, line 4. The gain is 4 million persons – a far larger amount. The reason that on net collective bargaining gains is that union membership is slight in the states that might enact legislation unfavorable to unions, while it is non-negligible in states likely to enact favorable legislation. Without specifying the specifics of laws and finding better ways to assess their impact (looking at the Canadian changes? Examining the effect of prevailing wage laws on density in construction? using the within state group variation, per Farber 2005), this suffices to make the point that if states were given authority for private sector labor law, the net effect on collective bargaining would likely be positive. 5. Conclusion 14 Federal law has failed to deliver workers the representation and participation they want. The law leaves millions of workers who want unions without workable ways to obtain union representation. It also restricts the ability of firms to set up nonunion committees or councile for workers to meet and discuss issues with management that many prefer to unions. The result is a huge unfilled demand for worker representation and participation in the US, both union and nonunion. If the most recent Peter Hart surveys are on target, that unfilled demand for unionism covers on the order of half of the nonunion work force in the country; while the unfilled demand for non-union workplace groups approaches 80% of the US work force (Freeman and Rogers, chapter 1). Efforts to reform the labor law have foundered. While there are no guarantees, turning the law regulating private sector labor relations and/or its enforcement to the states cannot do much worse than the US labor law is doing now. Washington has failed. Why not see if Sacramento and Bismarck, Albany and Oklahoma City, Des Moines and Detroit, Salt Lake City and Madison, ... can do better? 15 References Boston Federal Reserve New England Economic Review, June/July 1998 Brandeis, Louis, dissenting, New State Ice Co. v. Liebmann, 285 U.S. 262, at 311 (1932). Cimini, Michael “1982-87 State and Local Government Work Stoppages and their Legal Background” Comensation and Working Conditions, Fall 1998, pp 32-39 Clark, Jr. R. Theodore Public Sector Collective Bargaining: Past, Presetn, and Future http://www.npelra.org/articles/collective_bargaining.asp Donahue, John D. “Tiebout? Or Not Tiebout? The Market Metaphor and America's Devolution Debate” Journal of Economic Perspectives, Volume 11, issue 4, 1997 pp. 73_81 Farber, Henry 2005 Union Membership in the United States: The Divergence between the Public and Private Sectors WORKING PAPER #503 PRINCETON UNIVERSITYINDUSTRIAL RELATIONS SECTION September 2005 Farber, Henry S. 1988. \The Evolution of Public Sector Bargaining Laws." in When Public Sector Employees Unionize, Richard B. Freeman and Casey Ichniowski, eds., Chicago. University of Chicago Press. 129-166. Ferejohn John and Barry R. Weingast, eds., The New Federalism: Can the States be Trusted? (Stanford: Hoover Institution Press, 1997). Fraser Institute Economic Freedom of North America 2005 Annual Report (by Amela Karabegovi & Fred McMahon) http://www.freetheworld.com/efna.html Freeman, Richard B. and Robert G. Valletta. 1988(b). 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. 81-106. Freeman R and J. Rogers, “The Promise of Progressive Federalism,”in Making the Politics of Poverty and Inequality, ed. Jacob Hacker, Suzanne Mettler, and Joseph Soss (New York, NY: Russell Sage Foundation, forthcoming) Freeman R and J Rogers, What Workers Want, Cornell, 2 nd edition, forthcoming, 2006, Gottesman, Michael H.“Rethinking Labor Law Preemption” Yale Journal on Regulation 7 (Summer 1990): 355-410 Greve, Michael ” Laboratories of Democracy Anatomy of a Metaphor AEI Online (Washington) www.aei.org/publications/pubID.12743/pub_detail.asp. Krueger, Carl, State Collective Bargaining Policies for Teachers http://www.ecs.org/clearinghouse/37/48/3748.pdf Landau, Brent W. “State Employees and Soveriegn Immunity: Alternatives and Strategies for Enforcing Federal Employment Laws” Harvard Journal on Legislation, vol 39, pp 169-212 Political Economy and Research Institute, U Mass, Decent Work in America The State-by-State Work Environment Index, 2005 (by James Heintz, Jeannette Wicks-Lim, Robert Pollin) Saltzman, Gregory M. 1988. \Public Sector Bargaining Laws Really Matter: Evidence from Ohio and Illinois." in When Public Sector Employees Unionize, Richard B. Freeman 16 and Casey Ichniowski, eds., Chicago. University of Chicago Press. 41-80. Web sites with data www.unionstats.com/ www.nber.org/publaw 17 Table 1: The Distribution of Groups of State and Local Employees Covered by Different Public Sector Collective Bargaining Legislation, 1996 Type of Legislation # of State Employee Groups Percentage of State Employee Groups FAVORABLE* 146 63.8% Explicit Duty to Bargain 45 19.7% Implied Duty to Bargain 101 44.1% INTERMEDIATE** 31 13.5% Authorized not required CB 22 9.6% Meet and Confer/present proposals 9 3.9% 52 22.7% CB Barred 24 10.5% No Provision 28 12.2% UNFAVORABLE*** Source: Tabulated from Kim Ruebens, update of Valletta-Freeman data set, downloadable from NBER web site http://www.nber.org/publaw/. * 27 States Favorable to CB: Alaska, California.Connecticut Delaware Florida, Hawaii, Illinois Iowa, Maine Maryland, Massachusetts, Michigan, Minnesota, Montana, Nebraska, New Hampshire, New Jersey, New Mexico, New York, Oregon, Ohio Pennsylvania, Rhode Island, South Dakota, Vermont, Washington, Wisconsin ** 5 States Intermediate to CB: Arizona Indiana, Kansas, Missouri, W Virginia *** 18 States Unfavorable to CB: Alabama , Arkansas ,Colorado, Georgia, Idaho , Kentucky Louisiana ,Mississippi, Nevada, North Carolina, North Dakota, Oklahoma, South Carolina Tennessee, Texas, Utah, Virginia, Wyoming 18 Table 2: Statistical Analyses of Public Collective Bargaining Coverage, 2004 A. The level and range of Public Sector Coverage Compared to Private Sector Coverage States with favorable laws States with intermediat e laws States with unfavorable laws 48.6 (13.0) 28.0 (4.8) 22.2 (7.7) 22.9 to 72.5 23.0 to 33.6 10.4 to 39.6 Mean Density 9.4 (3.8) 8.9 (2.9) 5.1 (2.1) Min, Max Density 3.6 to 16.7 4.7 to 11.6 2.2 to 11.0 Mean Difference 39.2(10.7) 17.2(6.1) 19.0(3.1) Min, Max Difference 18.1 to 56.5 15.9 to 22.3 7.5 to 29.9 Public Sector Mean Density (SD) Range of Density (Min, Max) Private Sector Public Sector - Private Sector B. Coefficients and Standard Errors for the Regression of % Covered by Collective Bargaining in State on Independent Variables, 2004 State CB LAW Dependent variable constant Favorable Unfavorable 1 Percent Covered 28 20.6 (5.3) -5.8(5.5) 2.Percent Covered 10.9 18.5(3.8) 4.5 (4.1) 3.Ln % Covered 3.3 .52(.15) -.28(.16) 4. Ln % Covered 2.2 .50(.10) .07(.12) RTW Law %Covered Private Sector* R2 0.6 -4.6(3.2) 2.1(0.4) 0.8 0.6 -0.1(0.1) .55(.09) * in ln form in equations 3 and 4. Source: Coverage, www.unionstats.com/ Laws, Kim Ruebens update of Valletta-Freeman data set http://www.nber.org/publaw/). 0.8 19 Table 3: Number of States with Relevant Private Sector Labor or Social Policy Regulation in Eight Different Areas, by Public Sector Labor Relations Regulation Number of Laws ( Probability of Having the Law) All states 50 Minimum Wages Above Federal level Medicaid Eligibiity Criterion: Earnings > 100 Percent of Poverty State Support for Postsecondary Need-Based Aid 40% or more of Federal Pell Grant Aid Favorable 27 Intermediate and Unfavorable 23 16 (59%) 0 (0%) 13 (48%) 2 (9%) 11 (41%) 3 (13%) 10 (37%) 3 (13%) 23 (85%) 9 (39%) 11 (41%) 5 (22%) 10 (37%) 10 (43%) 21 (78%) 22 (96%) 115 53 16 15 14 Adopted Transitional Food Stamp Option for Families Leaving Welfare 13 Prevailing Wage Law for State Construction 32 State Earned Income Tax Credit 16 OSHA Complete State Plan 20 Job Quality Standards attached to Economic Development Incentives 43 Summary Statistic: Total number of favorable laws (average per state 168 3.4 Source: Laws, from Freeman and Rogers, 2006, Appendix. 4.3 2.2 20 Table 4: Number of private sector employees “at risk” of losing collective bargaining coverage in states that would enact unfavorable CB state laws compared to the number of workers who might gain coverage in states that would enact pro-CB state laws States with public sector labor laws unfavorable to collective bargaining States with public sector labor laws favorable to collective bargaining Total Private Sector Employment 31.5 million 63.9 million Number of Unionized Workers 1.2 million 6.3 million Number of Nonunion Workers 30.3 million 57.6 million Potential Loss of Union Members 0.5 million to 1.2 million Possible Gain in Union Members 4.1 million Source: Numbers of employees by union status, Estimated lost members in unfavorable law states, based on equation 4, table 2 Estimated gain in members in favorable law states, based on equation 4, table 2 The estimates assumed a change in the law would induce an ln increase/decrease in union membership of 0.5 ln points.