Labor Unions and the Great Recession

This article examines the impact of the Great Recession on the U.S. labor movement. After reviewing the classic industrial relations literature on the relationship between unionization rates and business cycles, we analyze historical union density trends. After documenting the relentless downward trend in the private sector from the early 1980s, with no apparent relationship to the business cycle, we analyze the negative impact of the political dynamic that unfolded in the wake of the Great Recession on public-sector unionism in sharp contrast to what took place during the Great Depression. We also explore the new forms of labor organizing that have emerged in the private sector, which have capitalized on the growing public concern about rising inequality sparked by Occupy Wall Street.

“The Great Depression invigorated the modern American labor movement,” the New Yorker economic columnist, James Surowiecki (2011), declared in January 2011. “The Great Recession has crippled it.” Although the economy has gradually climbed out of the crisis precipitated by the 2008 financial debacle, the U.S. labor movement indeed appears crippled. In 2015, only 11.1 percent of the nation’s wage and salary workers were union members, and the figure was an even lower 6.7 percent in the private sector—a pale shadow of unionism in the mid-1950s, when overall density (the proportion of workers who are union members) stood at about 35 percent (and higher in the private sector, although comparable figures by sector are not available for that period). Public approval of unions also fell dramatically with the financial crisis, reaching an all-time low in 2009, when the Gallup Poll found that only 48 percent of Americans approved of labor unions, down from 75 percent in the mid-1950s (Saad 2015). That historic peak in union strength and public support reflected two decades of unprecedented government support for collective bargaining and other policies designed to reduce economic inequality. Forged in the crucible of the Great Depression, the institutions associated with those New Deal policies have been deeply and deliberately eroded over recent decades. And progressive public policy initiatives to restore the strength of those institutions were conspicuously absent in the aftermath of the Great Recession. On the contrary, the political influence of anti-union forces—already formidable before the 2008 crash—has continued to mushroom ever since, while the inequality gap has continued to widen.

To be sure, the decline in union density long predated the 2008 financial crisis. Since at least the early 1980s, obituaries for the U.S. labor movement like Surowiecki’s have appeared regularly in both scholarly and journalistic writings. Private sector deunionization began in the late 1950s, but accelerated rapidly in the 1970s (see figures 1 and 2). Starting in the early 1980s, often seen as a turning point in the fortunes of organized labor, not only union density but also the absolute number of union members went into free fall (see table 1). As previous research has shown, deunionization contributed significantly to the rapid growth of inequality since the 1970s (Western and Rosenfeld 2011), and is also associated with the deregulation and financialization processes that took off in the same period (MacDonald 2014; Milkman 2013). Yet there has been surprisingly little attention to the impact of the Great Recession—the deepest economic downturn since the 1930s—on the labor movement. Was it the final nail in U.S. labor’s coffin, or could it presage some sort of reprise of the 1930s labor up-surge and the public policy shifts that accompanied it?

A useful starting point here is the classic industrial relations literature, which from its inception was concerned with the relationship between unionization trends and business cycles. A century ago, John R. Commons, the field’s founder, argued that unions tended to grow in times of prosperity and to decline during economic downturns (Commons 1918, 1:10–11). Shortly after World War II, the eminent labor economist John Dunlop (1948, 190–92) put forward an alternative claim, namely, that spurts in union growth—which by all accounts are the only way in which lasting density increases occur—followed major economic depressions as well as wars. In the latter case, Dunlop suggested, union growth was due to “the rapid rise in the cost of living and the shortage of labor supply relative to demand” that typically accompanies major military conflicts. But he noted that unions also grew in periods of “unrest,” during severe depressions like those in the 1890s and the 1930s, when labor organizing was tied to broader radical social movements.

Building on Dunlop’s theory, the labor historian Irving Bernstein (1954) argued that ordinary business cycles had little or no effect on unionization rates, and that union growth was spurred only by depressions “so severe as to call into question the very foundations of society” (316), such as the crisis of 1893 and the Great Depression. He noted that both those labor upsurges were marked by a lag effect: unionism grew only after the economy had begun to recover (in 1897 and 1933, respectively). Concurring with Dunlop that wars also stimulate union growth, Bernstein concluded that unions “have been the beneficiaries of disaster” (317). Although there is no evidence that Karl Polanyi (1944) influenced the thinking of Dunlop or Bernstein, some of their arguments are echoed in recent Polanyian analyses of the historical dynamics of global labor movements. Particularly influential here is Beverly Silver’s (2003) argument that global labor movement upsurges are an integral feature of broader social responses to historical waves of marketization like those that preceded the Great Depression. Laissez-faire deregulation, Polanyi theorized, stimulates countermovements from below that aim to decommodify labor, demanding protective legislation, unemployment insurance, and unionization (Silver 2003,17; see also Evans 2008; Burawoy 2010).

However, among U.S. industrial relations scholars and labor economists, the Dunlop-Bernstein perspective had fallen out of favor by the 1960s, when a new consensus view emerged resurrecting the original Commons hypothesis that union membership growth tracked business cycles. The literature of that era, anchored by elaborate multivariate analyses, acknowledged the influence of additional exogenous factors (such as political climate, public attitudes toward unions, and labor legislation) on union density, but the central focus was on business cycles. Thus, Orley Ashenfelter and John Pencavel (1969), in a much-cited analysis of the 1900-to-1960 period, concluded, “A period of increasing employment is favorable to successful organizing drives: the organizing funds of unions will be larger, and the potential union member is more receptive,” adding, “With the advent of union security agreements … increases in employment often lead automatically to upturns in union membership” (437).

Figure 1.

Private Sector Unionization Rates, United States, 1950–1981

Source: For unionization rates, see www.unionstats.com, “Table I. U.S. Historical Tables: Union Membership, Coverage, Density, and Employment, 1973–2015.” For recession dates, see National Bureau of Economic Research, “US Business Cycle Expansions and Contractions,” www.nber.org/cycles.html. Note: See note 2 regarding the absence of data collection in 1982 and the changes in data collection methodology that followed.

Figure 2.

Private Sector Unionization Rates, United States, 1983–2014

Source: For unionization rates, see www.unionstats.com, “Table I. U.S. Historical Tables: Union Membership, Coverage, Density, and Employment, 1873–2015.” For recession dates, see National Bureau of Economic Research, “US Business Cycle Expansions and Contractions,” www.nber.org/cycles.html. Note: See note 2 regarding the absence of data collection in 1982 and the changes in data collection methodology that followed.

A recent review of the literature on union density trends in the United States and other affluent countries similarly concludes that union growth is pro-cyclical: “Employment growth as well as price and/or wage inflation enhances membership growth … [but] a rise in unemployment tends to reduce union growth and density” (Schnabel 2013, 258). Our own perspective is closer to Bruce Western’s (1997), whose comparative analysis of OECD countries led him to conclude that the institutional context of unionism matters far more than business cycles. “In the right institutional framework, unions grow even through downturns in the business cycle,” Western argued. “Without favorable institutional conditions, unions are vulnerable to cyclical changes in the economy and grow only through extra-institutional strategies, such as strike action” (102).

Table 1.

U.S. Union Membership and Density, by Sector, 1983–2014

Source: www.unionstats.com, “Table I. U.S. Historical Tables: Union Membership, Coverage, Density, and Employment, 1973–2015.”

Note: Recession periods (shaded rows): July 1990 to March 1991, March 2001 to November 2001, December 2007 to June 2009.

In the United States, scholarly interest in the relationship of unions to the business cycle receded in the 1980s, as increased attention to union decline eclipsed the earlier preoccupation with explaining patterns of union growth.¹ But in the wake of the Great Recession, the first economic downturn that qualifies as sufficiently “severe” (in Bernstein’s terms) to merit comparison to the 1930s, and in the wake of four decades of neoliberal marketization, it seems apposite to revisit the classic questions about the impact of economic crises on union density and on labor movements generally. That is our purpose here.

We begin with an overview of recent density trends in the United States, examining both the private sector, where one would expect the effects of business cycles to be most apparent, and the more highly unionized public sector. The public sector is not explored in the earlier industrial relations literature, simply because when commentators from Commons to Ashenfelter and Pencavel were writing, public sector unionism was extremely limited. In fact, its expansion since the 1960s initially masked the precipitous decline of private sector unionism. However, the shifts in institutional conditions that affected union density after the Great Recession were particularly significant in the public sector.

Our analysis of U.S. union membership trends highlights the relentless decline in both absolute numbers and density since the early 1980s in the private sector, with no apparent relationship to the business cycle. The primary driver of this decline was a wave of concerted employer attacks on labor unions in highly organized sectors; a secondary factor was the limited organizing efforts on the part of unions themselves. Although there is some evidence that organizing efforts may have disproportionately declined in periods of recession, the overall deunionization trajectory for this period does not present a cyclical pattern. On the other hand, the political dynamic that unfolded in the aftermath of the Great Recession negatively affected public sector unions, which had maintained stable levels of density before 2008. And in the private sector, where density is now in the single digits, the center of gravity of the U.S. labor movement shifted toward new forms of organizing. Building on the momentum of the 2011 Occupy Wall Street movement, organized labor has begun to capitalize on growing public concern about inequality in a series of innovative campaigns.

U.s. union density trends and the great recession

Union density trends in the U.S. private sector since 1955 do not conform to the conventional wisdom that unionization rates fluctuate in relation to the business cycle (see figures 1 and 2). Instead, the period from 1955 to 2000 was marked by continual and indeed accelerating erosion in private sector union density, during periods of recession and expansion alike. Figure 2 reveals that in the twenty-first century a somewhat different pattern emerged. During the brief 2001 recession, private sector density at first remained flat, but this was soon followed by a resumed decline; similarly, during the Great Recession (2007 to 2009) union density initially increased slightly, and then the downward trend resumed in 2009. In any case, the pro-cyclical unionization pattern suggested in the earlier literature appears to be absent during the entire period shown in figures 1 and 2.

The number of private sector union representation elections held under the auspices of the National Labor Relations Board (NLRB) has also declined sharply since the 1970s (see figure 3). Unlike the density data, these election data do suggest a pro-cyclical pattern, with especially steep drop-offs in the number of representation elections during periods of recession. But there is no cyclical pattern in regard to the frequency with which unions are successful in winning NLRB representation elections: as figure 3 shows, since the early 1980s the union win rate has increased slowly and steadily, in periods of recession and growth alike.

Figure 3.

National Labor Relations Board Representation Elections and Union Win Rates, 1973–2014

Source: U.S. National Labor Relations Board, annual reports for fiscal years 1973–2003, available at: https://www.nlrb.gov/reports-guidance/reports/annual-reports (accessed September 16, 2016); U.S. National Relations Board, representation petitions for fiscal years 2004–2014, available at: http://www.nlrb.gov/news-outreach/graphs-data/petitions-and-elections/representation-petitions-rc (accessed September 16, 2016).

During the Great Recession the union win rate rose more sharply, although whether or not this is related to the economic downturn is far from clear. In recent years union leaders have become increasingly disenchanted with the NLRB process and have turned instead to alternative paths to union recognition; that is surely one major factor contributing to the long-term decline in the number of NLRB elections. This recent diversification in organizing strategies also means that the pro-cyclical pattern in the number of NLRB elections offers only a partial view of trends in new organizing. And unions have become increasingly strategic in their approach to NLRB representation elections, particularly since the early 2000s, when the Bush administration’s NLRB appointments led to a series of decisions that were especially hostile to organized labor (Liebman 2007; Farber 2014). The post-2008 increase in win rates, then, may simply reflect the Obama administration’s relatively union-friendly orientation (Scheiber 2015), rather than the impact of the economic crisis. In any case, the recent uptick in NLRB election win rates failed to stem the relentless decline in private sector union density, which has continued without interruption since 2008.

Apart from the effects (or lack thereof) of business cycles, many different factors can affect union density trends. Actively recruiting new members into the ranks of organized labor is the primary way in which unions themselves can act to increase the unionization level. But other forces that affect density are entirely beyond union control. All else being equal, if employment declines in a highly unionized sector of the economy, or expands in a nonunion (or weakly unionized) sector, union density will fall. Conversely, if employment expands in a highly unionized sector or declines in one where unionism is absent or weak, the overall level of union density will rise. In addition, labor market churning is an inherent feature of market economies, with new jobs constantly being created and old ones being destroyed.

This dynamic, along with the effects of population growth and labor turnover, means that simply to maintain union density at a given level requires a great deal of new organizing; to increase density requires even more extensive efforts. In the contemporary United States, where unionization is highly concentrated in “legacy” industries that in most cases are no longer growing, ongoing deunionization is virtually inevitable unless organizers can recruit massive numbers of new members in expanding industries. Although there have been some notable attempts in recent decades at such large-scale recruiting, particularly in the late 1990s, in the face of a hostile institutional environment most U.S. labor unions have concentrated instead on defending their past gains.

These dynamics are particularly salient in the private sector, but economic expansions and contractions also can affect public sector union density. Here, too, institutional factors play a crucial role. In the private sector, the U.S. political and regulatory context became increasingly unfavorable to unionism starting in the 1970s, but that was far less often true in the public sector until very recently. However, in the twenty-first century, and especially since the Great Recession, concerted political attacks on public sector unionism have spread, and anti-unionism has become a cornerstone of conservative ideology. Both reflecting and contributing to this development, public attitudes about unions are sharply polarized politically as well. In a 2015 Gallup poll (Saad 2015), for example, 55 percent of Democrats but only 18 percent of Republicans wanted unions “to have more influence.”

Table 1 presents detailed data on U.S. density trends for the years from 1983 to 2014, for both the public and private sectors.² It shows that both the absolute number of private sector union members and private sector density declined steadily, in periods of recession and growth alike, over those three decades. By contrast, in the public sector density levels were essentially stable until 2011, regardless of whether the economy was in recession or not. And until 2009, the absolute number of public sector union members increased steadily, apart from minor year-to-year fluctuations. That trend continued into 2009, thanks in part to the American Recovery and Reinvestment Act, signed early that year, which helped to sustain employment levels in many public-sector occupations.

However, in 2010, the absolute number of public sector union members began to decline, reflecting cutbacks in state-level public sector employment tied to austerity measures precipitated by the 2008 financial crisis. More than half the states laid off public employees in fiscal years 2010 or 2011, in efforts to reduce or eliminate budget shortfalls (National Governors’ Association and National Association of State Budget Officers 2010). State layoffs of public workers also occurred in some previous economic downturns, but to a far lesser extent than in 2010–2011 (Campbell and Sances 2013, 266–67). In the years just after the Great Recession, public sector job losses disproportionately affected African American and female workers, especially African American women (Laird 2015; Cohen 2015).

Changes in the absolute number of public sector employees do not necessarily presage shifts in union density, however, as they typically affect both the numerator (number of union members) and the denominator (number of workers) of the density ratio. But after the 2010 midterm elections brought Republicans into political power in many key states, austerity measures combined with a wave of direct political attacks on public sector collective bargaining to produce an unprecedented falloff in public sector density (Lafer 2013). Ironically, Wisconsin, where these political attacks were especially prominent, in 1959 had been the first state to pass legislation creating collective bargaining rights for public sector workers. The 2011 attack on public sector bargaining rights led by Governor Scott Walker sparked vigorous resistance and a dramatic political struggle, but ultimately Walker prevailed and the state suffered a precipitous loss in public sector union density (see figure 4).

Figure 4.

Public Sector Density Declines After Passage of the Restrictive Laws

Source: For unionization rates, see www.unionstats.com, “Table II. State: Union Membership, Coverage, Density and Employment by State and Sector, 1983–2015.” For list of states with restrictive laws, Lafer 2013.

Figure 5.

Public Sector Density Declines Before Passage of the Restrictive Laws

Source: For unionization rates, see www.unionstats.com, “Table II. State: Union Membership, Coverage, Density and Employment by State and Sector, 1983–2015.” For list of states with restrictive laws, Lafer 2013.

Similarly, Indiana, Michigan, and New Jersey experienced substantial erosion in public sector union density (see figures 4 and 5), although in these states that erosion (and the political attacks on public sector unions that drove it) had actually begun well before the higher-profile wave of attacks that emerged in 2011. On the other hand, in several other states that passed legislation restricting public sector collective bargaining rights in 2011 and 2012, the impact on union density has thus far been minimal or nonexistent (see figure 6). Reflecting these uneven state-by-state patterns, the overall decline in public sector union density nationwide has been relatively modest to date (see table 1).

In the private sector, unionization also varies widely among states; in 2014 density rates ranged from 2.5 percent in North Carolina to 14.9 percent in New York. Unionism in the United States is highly geographically concentrated: in 2014, seven states—California, New York,

Figure 6.

Public Sector Density Fluctuations in Other States with Restrictive Laws

Source: For unionization rates, see www.unionstats.com, “Table II. State: Union Membership, Coverage, Density and Employment by State and Sector, 1983–2015.” For list of states with restrictive laws, Lafer 2013.

Note: Does not include states where restrictive laws were passed but later overturned.

Illinois, Pennsylvania, Michigan, Ohio, and New Jersey—accounted for over half (53.8 percent) of all private sector U.S. union membership, although for only 36.3 percent of private sector employment (Hirsch and Macpherson 2015). Similarly, 53.5 percent of the absolute loss in private sector density between 2006 and 2014 was absorbed by those seven states. But there was also substantial variation among the seven: the largest private sector union membership losses were in Michigan, Illinois, Ohio, and New Jersey, while California was alone among this group of seven states in experiencing a modest rise in the absolute number of private sector union members. Three of these states—Illinois, Ohio, and especially Michigan—also suffered absolute losses in private sector employment during the 2006-to-2014 period, especially in manufacturing and construction, the industries most affected by the Great Recession. Employment fell in manufacturing in all seven states, with particularly sharp drops in Ohio, Michigan, New York, and New Jersey. Six of these seven highly unionized states also experienced absolute employment declines in construction (New Jersey was the only exception), with particularly steep declines in California and Illinois.

These state-to-state variations have larger implications. Higher unionization rates are associated with lower rates of poverty and inequality, in part because highly unionized states also tend to have politically influential labor movements that seek and often win improved minimum wages and other legislation that benefits union and nonunion workers alike. High unionization levels in particular states can also affect normative expectations for better jobs and more equitable wages (see Western and Rosenfeld 2011; Brady, Baker, and Finnegan 2013). State-level political attacks on private sector unions have been increasing across the nation, along with those focused on the public sector, although thus far Michigan is the only state in this group of seven that has passed right-to-work legislation.

Unionization is unevenly distributed not only geographically but also across industry groups. Moreover, as figures 4 and 5 suggest, as the economy recovered from the Great Recession, nonunion jobs often replaced union jobs. Not only did union membership decline sharply between 2006 and 2014 in the industries most affected by the economic downturn, but union membership fell faster than employment in all but one of those industries (see figure 7). The lone exception was the construction industry, a unique case both in regard to its central role in the Great Recession and also because building trades workers often retain their union affiliation even while they are unemployed. In manufacturing, wholesale and retail trade, information services, and the finance, insurance, and real estate industry group (FIRE), union membership declined far more than employment in this period. In “other services” and in the highly unionized education and transportation and utilities industries, although employment grew slightly in this period, union membership suffered a decline. Moreover, in those industries where employment did expand over this period, although union membership also grew, in most cases it lagged behind employment growth; the two exceptions were the accommodation and food services industry group and public administration.

Employment growth between 2006 and 2014 was greatest in industries with low union density (in 2006), whereas in industries with higher density, employment growth was weak or negative (see figure 8). In addition, union density declined over the 2006–2014 period in most of these industry groups (see table 2). The only ones where density increased were construction (for the reasons already suggested), accommodation and food services (where the increase was marginal), and public administration. Density was unchanged in professional services and health and social assistance. In the other eight industry groups shown, union density declined, with an especially steep drop in transportation and utilities.

The overall tendency for unionized jobs to be replaced by nonunion ones in the post-2008 recovery is also supported by table 3, which presents data for the nation’s twenty most highly unionized occupations. In 2014 those occupations accounted for 43.6 percent of all union members, but only 20.7 percent of total employment. Union density declined in most of these occupations between 2006 and 2014, although it remained stable among construction laborers (for the reasons noted) and rose among registered nurses, police officers, secretaries and administrative assistants, and bailiffs, correctional officers, and jailers. Even in states where public sector collective bargaining laws came under attack, police were often exempted; registered nurses and corrections are among the rare fields in which unionization has been strengthened in recent years, a trend that remained intact in the aftermath of the economic crisis.

Figure 7.

Changes in Employment and in Union Membership Between 2006 and 2014, Selected Industry Groups.

Source: www.unionstats.com, “Table IV. Union Membership, Coverage, Density and Employment by Industry, 1983–2015.”

On the whole, however, unionization continued its relentless long-term decline during the Great Recession. Union density declined in part because new organizing did not keep up with labor force growth, especially in expanding industries and sectors. In addition, employers in many industries actively sought to undermine or eliminate unions in sectors that historically had been highly organized. The result was that growth in nonunion jobs generally

Figure 8.

Change in Total Employment, 2006–2014, and Union Density in 2006, Selected Industry Groups

Source: U.S. Current Population Survey, Outgoing Rotation Group, 2006 and 2014.

Table 2.

Union Density in Selected Industry Groups, 2006 and 2014

Source: U.S. Current Population Survey, Outgoing Rotation Group, 2006 and 2014.

Table 3.

Employment, Union Membership, and Union Density in Selected Occupations, 2006–2014, Ranked by 2014 Union Membership

Source: Hirsch and Macpherson 2012, 2015.

outpaced that in union jobs, a long-term trend that was also a salient feature of the post-2008 recovery.

An institutional perspective

The short-term effects of the Great Recession were far less consequential for unions than the institutional environment, which had been hostile to organized labor for the previous three decades, especially in the private sector, and which continued to deteriorate in the wake of the economic crisis, in sharp contrast to what took place in the Great Depression. Since the late 1970s, employers had become increasingly adept at manipulating the NLRB election process with a variety of tactics that create new obstacles for unions. In the same period, employers also began routinely using “replacement workers” in economic strikes, greatly reducing the effectiveness of what had previously been a key source of union leverage (Rhomberg 2012). Tactics and strategies vary across industries, but since the late 1970s employers have consistently been on the offensive, acting both to limit the spread of unions into new sectors and to weaken the power and influence of those established unions that have managed to survive (Logan 2006, Bronfenbrenner 2009).

In the aftermath of the 2008 financial crisis, a new wave of political attacks on organized labor emerged. But this was not simply a continuation of employers’ previous anti-union efforts: for the first time the public sector became a central battleground. Here the attacks did not come directly from employers—which were government agencies—but rather from political actors. Attempts at undermining public sector unions had appeared periodically before the Great Recession, although they were not related to the business cycle. However, such efforts accelerated enormously in the wake of the 2008 crash and the state and local budget deficits that it helped to create. “Public sector labor law has long been subject to changes,” the legal scholar Joseph Slater (2013) concluded after reviewing developments from the 1980s onward. “Nothing, however, prepared those in the field for what was going to happen in 2011” (526).

The attacks on public sector unions that emerged that year were nakedly political in character. The ties between organized labor and the Democratic Party, and unions’ longstanding tradition of generously supporting Democratic electoral candidates had for decades made unions anathema for the political right. Now anti-union organizations such as the American Legislative Exchange Council (ALEC) capitalized on the unique opportunity presented by a series of state-level Republican electoral victories in the 2010 midterm elections. In 2011 and 2012 alone, fifteen states passed laws restricting public employees’ collective bargaining rights (although three of these were later overturned in popular referenda). ALEC wrote model legislation and disseminated it to sympathetic elected officials in various states, an approach that proved highly effective. Private sector unionism already had declined dramatically; by contrast, in the public sector organized labor’s strength remained intact. Thus the unions in that sector were the target.

Although this focus on the public sector was the distinctive feature of the post-2008 period, anti-union efforts continued to escalate in the private sector as well. ALEC and other right-wing groups promoted proposals for “right-to-work” legislation, which prohibits collective bargaining agreements to require that all covered workers pay union dues. Right-to-work laws were introduced in nineteen states in 2011 and 2012, and were soon passed in three former union bastions in the Midwest: Indiana in 2012, followed by Michigan in 2013 and Wisconsin in 2015 (Lafer 2013). ALEC also promoted legislation on a variety of other labor matters, including bills eliminating New Deal–era “prevailing wage” laws. These laws require firms with public contracts to pay the wages and benefits that the majority of workers in a region and occupation receive (typically the wage and benefit level specialized in collective bargaining agreements). Indiana passed legislation eliminating one such law in 2015 (Lafer 2013, Davey 2015). Nine other states already had repealed their prevailing wage laws in the 1980s, but efforts to accomplish this in other states were ramped up in the aftermath of the Great Recession.

By any standard, then, the institutional context in which unions are struggling to survive, which had become increasingly hostile to organized labor over the three decades before the Great Recession, seems even more treacherous in the post-2008 period. This presents a striking contrast to what took place during the 1930s, when the economic crisis was the impetus for a set of major policy breakthroughs in support of collective bargaining and social protection for workers. (Public sector unionism barely existed in that period, so it was simply not part of the equation.) Many of the recent anti-labor legislative proposals put forward by ALEC and other such groups are explicitly designed to undercut or eliminate those New Deal–era protections.

This raises the question: Why are the outcomes after 2008 so different from those in the 1930s? Why did a labor upsurge like the one that developed during the Great Depression not recur in the wake of the Great Recession? Why, instead, does union density continue to stall or decline today? We cannot definitively answer this question, but we can suggest some possible explanations.

Although employer opposition to unions was virulent in the 1930s, just as it is today, there are some differences worth noting. First, President Franklin D. Roosevelt took a bold stand in support of unions and collective bargaining, and he had the political leverage to make that a reality. Obama has taken much weaker positions in defense of unions, and has faced a sharply divided Congress, with one house dominated by elected officials who are notoriously anti-union, all of which greatly constrains his ability to act on even his relatively modest efforts to support organized labor. Second, although employers strongly opposed unions in the 1930s, the depth of the crisis meant that many of them were directly dependent on the federal government to restore stability to the overall economy and to address the problem of cutthroat labor market competition. In this period, a large fraction of the employer community understood the political necessity of agreeing to concessions in regard to collective bargaining in order to save capitalism from itself. By contrast, after the 2008 crash, Congress moved relatively quickly to bail out the banks with no strings attached, and demanded virtually no concessions from employers in return. Once the bailout was accomplished, employers were far less dependent on the administration or on congressional legislation than their counterparts in the 1930s to get the economy running again (although some did benefit from the 2009 American Recovery and Reinvestment Act).

Another possibility is that it is too early to draw definitive conclusions as to the impact of the two crises on labor. The upsurge in union membership occurred several years after the 1929 stock market crash. Although it does not seem likely that anything comparable to the giant upsurge that followed the passage of the National Labor Relations Act of 1935 will recur, that sort of outcome cannot be ruled out entirely. We may just now be seeing the initial fruits of a post-recession organizing upsurge. In 2014 and 2015 union density inched up slightly in a few states and industries (Milkman and Luce 2015; Bureau of Labor Statistics 2016a). The number of large strikes, and the number of workers involved in strikes, also increased slightly in 2015 (Bureau of Labor Statistics 2016b). Moreover, alongside this renewed activity on the part of traditional unions, the labor movement’s strategic orientation has begun to pivot, largely in response to the anti-union political climate and employer attacks. Unions and labor movement allies alike are exploring alternative organizational forms that in some respects recapitulate the strategic repertoire of the pre-1930s labor movement (see Milkman 2006, 2013). It is to those developments that we now turn.

New directions: “alt-labor” and twenty-first century u.s. unionism

The besieged organized labor movement began to experiment with a variety of new initiatives in the 1990s and early 2000s. In the late 1990s the AFL-CIO launched an energetic campaign designed to “organize the unorganized,” followed a decade later by high-profile efforts to win labor law reform. Several large unions left the AFL-CIO in 2005, forming a rival federation designed to reposition the unions to better cope with the increasingly hostile institutional environment, and again with a professed commitment to expanding new organizing. Although these efforts led to some successful union drives, they were typically modest in scale and thus insufficient to reverse the relentless decline in union density. Other campaigns, such as the effort to pass the Employee Free Choice Act, failed entirely to achieve their goals. Unionism continued to decline, and the situation became increasingly desperate as political attacks on collective bargaining ramped up after the Great Recession.

One response to this dire situation came from a variety of nonunion community-based labor organizations, many of which have come to be known as “worker centers” (Fine 2006) and more recently as “alt-labor” (for “alternative labor”) organizations. These groups. which unlike traditional unions were expanding during the 1990s and 2000s, began to experiment with new approaches to worker organizing. Focused especially on immigrants and others concentrated at the very bottom of the labor market, these organizations emerged as highly effective advocates for low-wage workers. They succeeded in calling media and public attention to labor and employment law violations, such as payment below the legal minimum wage (or in some cases outright nonpayment), and won legal remedies for some of the victims. Initially the traditional unions were deeply skeptical of the effectiveness of these approaches, but that gradually changed in the 2000s as more and more individual unions as well as the AFL-CIO began to partner with worker centers.

There were two key turning points in the labor movement’s shift in this direction, one before and one soon after the Great Recession. The first came in 2006, when massive immigrant rights marches swept the nation in response to the threat that a draconian immigration bill, “The Border Protection Anti-Terrorism and Illegal Immigration Control Act of 2005” (H.R. 4337), which passed the U.S. House of Representatives in late 2005, would become the law of the land. Millions of immigrants demonstrated against this legislation in the spring of 2006, mobilized by a coalition of worker centers and immigrant rights groups, with support from some labor unions as well. Immediately afterward, formal partnerships between the AFL-CIO and worker centers began to develop, a process that has continued ever since. These dramatic immigrant rights protests dispelled any lingering doubts in the House of Labor about the “organizability” of low-wage immigrant workers; at the same time the achievements of the worker centers were winning growing respect from traditional unionists.

The second turning point was the Occupy Wall Street uprising, whose meteoric rise in the fall of 2011 transformed the national political debate and raised public awareness about the growing inequality between rich and poor. Organized labor’s previous efforts to call attention to rising inequality had failed to gain any traction with the public, but after 2011 unions began to build on the foundation established by the Occupy movement. The most dramatic example is the Service Employees International Union (SEIU)’s organizing campaign in fast-food chains, demanding wages of $15 per hour and the right to unionize. Launched in New York in late 2012, this “Fight for 15” has since spread nationwide, propelled by a series of one-day demonstration strikes that garnered extensive publicity. Although sponsored by a traditional labor union, the SEIU, this campaign essentially has adopted the strategic repertoire of the worker center movement. There is no immediate prospect of union recognition, but the effort has succeeded in shining a bright light on the low wages and other workplace abuses in this industry. The strikes have attracted support from workers in other low-wage sectors: in December 2014, April 2015, and November 2015, airport workers, domestic workers, convenience store employees, and adjuncts joined in the one-day “Fight for 15” strikes (for a fifteen-dollar hourly wage). Another worker-center-like union-sponsored campaign is OUR (Organization United for Respect) Walmart, which launched a series of “Black Friday” strikes as well as other efforts to put pressure on the nation’s largest employer to improve its employment practices.

The fifteen-dollar-an-hour demand first floated by the fast-food organizing effort also sparked campaigns to raise the minimum wage in key cities and counties where unions still have a strong presence. Seattle and SeaTac in Washington State, along with others, including San Francisco, Emeryville, and Los Angeles in California, have passed laws that will raise the overall minimum hourly wage to fifteen dollars or higher in the coming years, and advocates have won more modest increases in minimum wages in over a dozen other cities and states across the nation. These highly successful campaigns built directly on the Occupy movement’s successes in raising public awareness of inequality. In 2014 alone, fourteen states raised their statewide minimum wage (Luce 2015). Public approval of unions has also increased, climbing back from its low point in 2009 (Saad 2015).

There are no systematic data available on the scale of these alt-labor efforts, but we can sketch the trajectory of their recent growth by drawing on a variety of published accounts. There were 137 worker centers in the United States in 2003 (Fine 2006), and 160 in 2007 (Fine 2007, 57). The number grew substantially in the aftermath of the Great Recession, to over 200 by 2010 (Fine 2011, 615) and, according to one recent estimate, to a total of 230 by 2013 (Narro 2013). In addition, several worker centers that began as local operations have expanded into national operations. These include the Restaurant Opportunities Centers United (ROC-United), with 11 local organizations across the nation; the National Domestic Workers Alliance, with 42 affiliates; and the National Day Laborers Organizing Network, with 43 affiliates (Fine 2015, 17). Other examples include the Food Chain Workers’ Alliance (with which ROC-United is also affiliated) and the National Guestworker Alliance. Although there are no reliable estimates of how many workers are affected by these efforts, many of which are modest in size, the Food Chain Workers’ Alliance alone claims to represent 300,000 workers. Many of these organizations also have successfully litigated claims of wage theft and won back pay for substantial numbers of workers.

We can be a bit more precise in estimating the numbers of workers who have benefitted from recent increases in the minimum wage. These are summarized in table 4, which suggests that over 2 million workers have been impacted by these new laws at the local level from 2012 to early 2016 alone. Table 4 also includes data on several major employers who have announced wage hikes in response to the “Fight for 15” and related campaigns, which we estimate to have benefitted almost 1 million workers. Table 4 does not include the statewide minimum-wage increases that have been legislated in the aftermath of the Great Recession; an Economic Policy Institute study estimates that those laws had affected 3.1 million workers by 2014 (Cooper 2014).

Table 4.

Estimated Number of Workers Impacted by Minimum Wage Increases and Employer-Sponsored Wage Hikes, 2012–2015

Source: Authors’ compilation.

Notes: Estimates of workers impacted come from a variety of sources. Where possible, we used city studies, academic reports, or newspaper articles that estimated the impact, or public company statements announcing wage increases. In cases where these sources provided a range of potential workers covered, we used the midpoint. In a few cases, we calculated our own estimate of potential impact extrapolating from data for cities similar in population size. For further details on this methodology, please contact the authors.

^aThe Long Beach, Calif., ordinance applies only to hotel workers; a citywide minimum wage is under consideration.

^bAuthors’ estimate of workers impacted, based on the size of the labor force and coverage data from similar cities.

^cGWashington, D.C., passed an initial citywide minimum wage in 1993 and then raised the wage rate in 2013. San Francisco passed an initial citywide minimum wage in 2003; voters then passed an increase in the wage rate in 2014.

States have set and raised their minimum wages for decades, and increases often come in waves (examples include 2004 and 2006), but local minimum-wage laws are a relatively recent phenomenon. As of 2010, only a handful of cities had passed their own minimum wage ordinances, but between 2012 and early 2016, thirty-two municipalities did so. In September 2015, seven cities in the San Francisco Bay Area announced plans to work together to establish a regional minimum wage, another recent innovation. Organized labor has also promoted a variety of legislative measures at the state and local levels aimed at improving the situation of low-wage workers, mandating benefits such as paid family leave and paid sick days, and improving enforcement of labor standards. (Milkman and Appelbaum 2013; Reich, Jacobs, and Dietz 2014).

Both these legislative initiatives and the spurt of alt-labor organizing efforts resemble pre–New Deal labor movement strategies, in contrast to the NLRB-based union organizing campaigns that became common in the mid-twentieth century (see Milkman 2013). In the Progressive Era of the last century, labor reform groups and their middle-class allies publicized sweatshops and employer abuses and provided educational and social services for immigrant workers in much the same way that worker centers do today (Flanagan 2002; Stromquist 2006). These reformers also promoted unionization and campaigned for progressive legislation, including the first state minimum-wage laws (although at the time these applied only to women and children). In that era the cities and states were “laboratories of democracy,” whose reforms helped to set the stage for the landmark labor and employment legislation of the New Deal.

In the absence of systematic data, we can only sketch the achievements and prospects of the new alt-labor efforts and the flurry of legislative activity that have emerged in a fragmentary way. But they stand out as central components of organized labor’s response to the Great Recession. To be sure, these developments have had no apparent effect on the steady decline in union membership and density. But like the parallel efforts a century ago, they could represent the embryo of a Polanyian counter-movement in response to the post-1970s wave of neoliberal marketization.

To sum up: Although the Great Recession does not appear to have been a significant factor in the long-term decline in union density, which had been under way for many decades, the 2008 financial crisis nevertheless did impact the labor movement. It opened up political space for right-wing attacks on the rights of unions to exist and bargain collectively in both the private and public sectors, further eroding an already hostile institutional environment. Steep job losses in traditionally unionized industries during and after the Great Recession added to the challenge for the labor movement in maintaining union density. Nevertheless, the surge in successful efforts to increase state and local minimum wages and the growing momentum of innovative organizing efforts like the “Fight for 15” offer some hope that James Surowiecki’s (2011) obituary for the labor movement may prove to have been premature.