In the American political system, access to key policymakers has long been recognized as a vital prerequisite for obtaining influence over policy decisions. “Toward whatever institution of government we observe interest groups operating, the common feature of all their efforts is the attempt to achieve effective access to points of decision,” Truman ([1951] 1971, 264) wrote in The Governmental Process. Of course, access alone does not guarantee influence over decisions, and can even provide an illusory indicator of the actual power of an interest group (Frymer 2010; Johnston 2015; Lowery 2013). Nonetheless, interest group leaders and lobbyists have remained adamant about the importance of securing access to decision-makers in American government, including not only members of the federal legislature, but also members of the executive branch (Ainsworth 2002, 131; Nownes 2006, 90; Baumgartner et al. 2009, 154). There is abundant evidence—today as when David Truman wrote over a half-century ago—that groups urgently seek to be at the table when issues are discussed and decisions made.

It is desirable, therefore, to develop a better understanding of how political access is established, and how and why it fluctuates in response to changing circumstances (Baumgartner and Leech 1998, 177–180; Hansen 1991). This article examines how access of labor unions to the president of the USA has been affected by the growth of partisan polarization since the 1960s. As a massive literature shows, the two parties have become more internally unified and less regionally distinct, with liberals in all parts of the country finding a home in the Democratic Party and conservatives joining the Republican Party (Thurber and Yoshinaka 2015). The result is an altered environment for interest group action (Sinclair 2006, 308–323; Karol 2015). Groups that achieved access to both parties in the past may no longer succeed under conditions of polarization, regardless of the wealth or resources they command. Likewise, old strategies of navigating between the parties, often embodied in an official group ideology of “nonpartisanship” (Rogin 1987), are challenged by the extension of partisan conflict to a growing number of issues (Layman and Carsey 2002).

While the effects of polarization on interest group access and strategy are clearly important, they have only rarely been explored. A powerful way to study these effects more closely is a case study, defined “as an intensive study of a single unit for the purpose of understanding a larger class of (similar) units” (Gerring 2004, 342). A particularly useful case is that of organized labor, which was very active in the national political arena before polarization developed (making its first endorsement in a presidential campaign in 1908) and has remained quite active in the period since (Greenstone 1977; Francia 2006). Because of this long-lasting involvement, there exists an abundance of primary material, as well as a rich secondary literature, that documents union political activity, and allows confident and nuanced historical comparisons.

Political access can be defined in straightforward fashion as “the ability to see and speak with policymakers” (Nownes 2001, 228). To a considerable extent, the concept is easily operationalized by simply recording the frequency, location, duration, and, if possible, content of meetings and conversations in which policymakers and interest group representatives interact.

More complicated is the question of “access to whom?” Interest groups usually seek influence at numerous institutional locations within the federal government, including parts of Congress, the judiciary, the executive branch bureaucracy, and the White House itself (Truman 1971, 264–270). Although any of these access points can, depending on circumstance, prove determinative for a group’s goals, the primary focus here will be on access directly to the president of the USA. For some groups, this choice would be far too narrow and limiting, since regular access to the president is highly difficult to achieve and rarely offered (Nownes 2006, 68). In the case of unions, however, it is appropriate. Top union leaders, and especially the head of the main labor federation (since 1955, the American Federation of Labor–Congress of Industrial Organizations, or AFL–CIO), have regularly sought and achieved access to the president, both in meetings at the White House and other locations, and through phone conversations (Dark 2001; Dubofsky 1994; Minchin 2017; Robinson 1981).

To better allow comparison, the time frame for this study includes the periods before and after the development of polarization. The onset of polarization did not occur suddenly, of course, but developed over a period of years. While the exact dating of a transitional period could be debated, most observers would agree that the arrival of partisan polarization (as opposed to political conflict in general) had not become obvious in the 1970s (when political scientists were obsessed with party decline, not revival). And few would dispute that it had become readily apparent by the late-1990s (Brownstein 2007; Sinclair 2006; Mann and Ornstein 2016). Based on this rough demarcation, we can place the presidents prior to the late-1970s in a pre-polarized age, and those after the 1990s in a time of full polarization. Accordingly, the presidencies of Lyndon Johnson and Richard Nixon, a successive Democrat and Republican pair, can serve as examples of union presidential access in the period before polarization. The presidencies of George W. Bush and Barack Obama, likewise a successive Republican and Democrat pair, can exemplify the nature of union access after the emergence of polarization. In their times, these four presidents were considered relatively conventional politicians who were quite representative of the mainstream of their parties (in other words, they were not outsiders elected though a fluke of the electoral process). The comparison of how each managed union access can help capture the effects of partisanship within each period, as well as the effects of polarization between the two periods.

Naturally, the rise of polarization is not the only contextual development of significance that separates these periods. Of great importance is the phenomenon of labor union decline, especially in the private sector. Since the middle of the twentieth century, private sector unions have seen their collective bargaining power weaken markedly as a result of changes in global markets, deindustrialization, employer hostility, new management techniques, and an ossified and unfavorable legal regime (Estlund 2002; Rosenfeld 2014; Friedman 2009). Overall union density (a term for the percentage of the workforce unionized) stood at a mere 10.7% in 2016, down from a high of about 35% in the mid-1950s, and in the private sector alone, density declined to 6.4%—lower than before the arrival of the New Deal (Bureau of Labor Statistics 2018). Unions in the private sector have declined so severely that their collective bargaining agreements now lack significant macro-economic consequence, and strikes—traditionally useful as both negative and positive inducements when dealing with politicians—have become quite rare. Public sector unions, meanwhile, insulated from many of the pressures that have devastated their private sector counterparts, have fared much better, maintaining a 34.4% density rate in 2016. The relative health of public employee unionism has helped to provide some resilience in the absolute number of union members (there were a total of 14.8 million in 2017 compared to 17.7 million in 1983) (Bureau of Labor Statistics 2018). The diverging fortunes of private and public unionism have resulted in about half of union members in the USA being employed by government, not private industry.

The downward direction of union density has encouraged widespread claims about declining union power, typically based on the assumption that a decline in density brings a corresponding diminution in overall union resources (Andrias 2016; Lichtenstein 2002, 2011; Rosenfeld 2014). Valuable union assets typically include total membership (the basis for voting, volunteering, and other forms of political behavior), money, staff, specialized information, and even collective bargaining capacity (with union decisions regarding strikes and contracts serving as potential bargaining chips). If these resources decline, it is reasoned, so will access. Inevitably, unions will lose their “competitive advantage” in politics to other groups and organizations more capable of meeting the needs of elected officeholders (Hansen 1991, 13–17). This resource-based explanatory approach generates an unambiguous prediction: union access will be greatest during the mid-twentieth century, when unions had larger numbers and greater centrality in industrial relations, and will diminish in linear fashion for presidents of both parties from that highpoint. The analysis here takes a different approach, questioning the degree of union resource decline and, more importantly, placing labor’s role in the context of partisan change. The dynamics of polarization, it is argued, have produced incentives driving unions and Democratic presidents closer together, and unions and Republican presidents further apart, regardless of the size and utility of union resources.

From insider bargaining to partisan mobilization

For much of American history, the parties were notorious not for their polarization, but for their internal heterogeneity and lack of ideological coherence. Especially following the growth of southern Democratic opposition to the New Deal in the late 1930s, both parties were divided along regional and ideological lines, with the question of union power often a key factor in spurring internal conflict (Farhang and Katznelson 2005; Sinclair 2006, 71; Miller and Schofield 2008). Southern Democrats regularly joined conservative Republicans to oppose liberal goals, especially pro-union policies that threatened the configuration of class and racial power in their home region (Lichtenstein 2002, 110–114; Weatherford 2014). Yet, the national party also contained northern and western liberals with more favorable views of the role of government and, frequently, pro-union policy stances. The resulting coalition was, as James Campbell observes, “a fragile, unsettled, and ultimately unsustainable coalition of antagonistic voting blocs” (Campbell 2016, 154). The Republican Party had its own divisions as well, between liberals and moderates (largely from the northeast and west), often willing to forge compromises with organized labor, and conservatives (mostly from the Midwest and rural areas), who were unyielding in their opposition. The persistence of these internal partisan divisions meant that the passage of legislation required the construction of bipartisan coalitions. As journalist Ronald Brownstein put it, the mid-century era was an “age of bargaining.” It was a time, he observes, in which “almost every important decision in Washington required intense consultation and negotiation between and within the parties” (Brownstein 2007, 58).

Organized labor’s place in American politics during the age of bargaining reflected these distinctive partisan alignments. Since the early 1900s, labor had clearly been closer to the Democratic Party, with the American Federation of Labor’s first presidential endorsement going to William Jennings Bryan in 1908 (Greenstone 1977, 31). The passage of pro-union legislation during the New Deal era served to deepen labor’s ties to national Democrats. Still, unions found it possible to work with many Republicans in Congress and state and local governments, and most union leaders emphasized their “independence” from either party (Rosenfeld 2014, 160–161; Wilson 1979, 36–37). The building trades, in particular, were willing to endorse moderate Republicans, and eschewed the more liberal or social democratic orientation common among industrial unions. Thus, when Dwight Eisenhower became president in 1953 after 20 years of Democratic tenure, few found it surprising that he maintained a cordial relationship with the AFL–CIO and appointed a former president of the plumbers’ union as his secretary of labor (Dubofsky 1994, 210; Robinson 1981, 203). Eisenhower even consulted with union leaders in an effort to develop a set of labor-supported revisions to the Taft–Hartley Act—an industrial relations measure passed by Congress in 1947 over vehement union opposition (Weatherford 2014; Yeselson 2017). Although Eisenhower’s reform plan was ultimately abandoned in the face of congressional opposition, this overture (and others) was likely on the mind of AFL–CIO President George Meany when he stated in 1955: “We absolutely refuse to allow ourselves to be an appendage of the Democratic Party” (Wilson 1979, 36). Meany’s comment belied, of course, the integrated—even symbiotic—relationship that many union organizations had constructed with Democrats in Congress and in cities and states (Greenstone 1977). Fittingly, in a separation of powers system and federal republic, the relationship was complicated, and varied according to location and circumstance.

The age of bargaining, according to Brownstein was followed by an “age of transition,” triggered in part by the Democratic Party’s support for civil rights legislation in the mid-1960s. From that time onward, southern whites migrated into the Republican Party, and northern liberals moved into the Democratic coalition. As the ideological overlap of the two parties diminished, partisan disputes in the Washington community became increasingly contentious, angry, and uncivil. In Congress, party unity scores and other quantitative measures of polarization rose dramatically, fueled by the alignment of cultural, geographic, racial, ideological, and religious cleavages with partisan affiliations (McCarty et al. 2016; Campbell 2016). Of the two parties, it appeared that the Republican had moved furthest from the political center, as its congressional wing demanded radical changes in public policy and nurtured an uncompromising anti-government (and typically anti-union) zealotry (Mann and Ornstein 2016; Hacker and Pierson 2005). But the Democrats moved off-center as well, more decisively on racial and cultural issues, however, than on economic and welfare state policy (Sinclair 2006, 66; Hacker and Pierson 2011). For both parties, the path to legislative success became the intense mobilization of their own members rather than the forging of complex bargains among centrist lawmakers (Mann and Ornstein 2016).

As polarization intensified, the place of interest groups in national politics was affected as well (Sinclair 2006, 308–325). Parties increasingly sought to mobilize groups as agents of the party’s own programmatic agenda. For their part, groups found their allies ever more concentrated on just one side of the partisan divide, forcing them to choose one of the two parties as the primary focus of their political allegiance. Polarization thus cleaved the interest group community itself, leading groups to align decisively with one of the two parties. “Staying neutral has become a progressively more untenable strategy for major interest groups,” Sinclair observes (2006, 309). Under such conditions, John Mark Hansen (1991, 226) writes, “The party system dominates the pressure system.” The causal process is straightforward: “Where party cleavages reproduce interest group cleavages, lobbies march in lockstep with political parties” (Hansen 1991, 223). One consequence is that the conditions of presidential access, for labor and other groups, are altered, with partisanship taking on greater importance. The more intense polarization becomes, the more likely it is that access will be determined by partisan affiliation.

Union access to the president

Since the early twentieth century, unions have arguably been obsessed with maintaining access to presidents, especially those from the Democratic Party (Dark 2001, 69–73; Buhle 1999; Goulden 1972). It is through access that policy preferences can be communicated to the president and commitments secured to support union interests (Nownes 2006, 214). A place at the table when key issues are being discussed enables union leaders to influence an administration’s legislative strategy, including the crucial issue of the sequence and timing of agenda items (Johnston 2015). In addition, union leaders often see access to the White House as a means of consolidating their own organizational leadership position vis-a-vis possible internal rivals. Presidential access demonstrates the effectiveness and status of a leader, especially within a labor federation that has as its major purpose political action. And of no small importance, unions value high-level access because it establishes the legitimacy of their role as socio-political actors—a status that has been at serious question for much of American history. For all these reasons, the status of presidential access provides a valuable indicator of union influence during different periods.

Lyndon Johnson, Richard Nixon, and the age of bargaining

It is widely acknowledged that during his time as president, Lyndon Johnson had a productive relationship with labor unions, one that involved working closely on a wide range of issues and policies (Goulden 1972, 336–370; Robinson 1981, 233–250; Zieger et al. 2014, 232–233). Less recognized, though, is the extent to which this amicable relationship was something of a surprise. Johnson was, after all, very much a product of the conflicted Democratic Party of the New Deal era. While an admirer of Franklin Roosevelt, Johnson had also made his peace with the conservative anti-union forces that dominated Texas politics—an accommodation that produced a decidedly mixed congressional voting record on labor issues (Zelizer 2015, 70). Many union leaders thoroughly distrusted Johnson, and strongly opposed his selection by John F. Kennedy to be his vice-presidential running mate in 1960 (Robinson 1981, 219; Goulden 1972, 297–301; Puddington 2005, 45). Yet, as president, Johnson not only granted access to unions, but reached out on his own accord to establish close ties with the top leadership.

Shortly after the November 1963 funeral of President Kennedy, Johnson made telephone calls to AFL–CIO President George Meany and Walter Reuther, the influential leader of the United Auto Workers, promising to continue and deepen the policy agenda of his slain predecessor (Robinson 1981, 234; McMullan 1963). Johnson soon met with Meany personally and, as Meany recalled, “talked about labor, the Labor Department, and different things. He indicated that he was very anxious to go along with a list of things that we were for: federal aid to education, certainly the civil rights business” (Robinson 1981, 234). Johnson’s contacts were the beginning of a systematic effort to solicit union support, which was important not only in the legislative labyrinth on Capitol Hill but also in the Democratic Party’s presidential nominating process. In a meeting on December 5, 1963 with the AFL–CIO Executive Council (the Federation’s main governing body, composed of presidents of the major national unions), Johnson stressed: “I want you to know that the doors to this house will always be open to you for your ideas and problems. And the doors of the Cabinet officers, too” (McMullan 1963).

Johnson kept his word in the years that followed. “Johnson courted Meany more fervently than had any President,” a biographer of Meany wrote (Goulden 1972, 338). Meany himself recalled: “We were in constant contact with him. Every week there would be two or three telephone calls and visits…I was in the White House sometimes two or three times a week. I’d want to see him, or he’d call me” (Meany 1969). Johnson’s focus on legislative success was a major reason for his interest in labor, Meany stated. “He was quite aware of the fact that on some of these liberal measures where a few votes were needed to finally enact the legislation, our influence…could pick up votes that even he couldn’t pick up as President” (Meany 1969). The AFL–CIO’s chief lobbyist, Andrew Biemiller, also emphasized the legislative utility of an alliance with labor, and remarked that “there wasn’t any question that our relations with Lyndon were so good as to be almost incredible” (Robinson 1981, 244). The UAW’s Reuther, generally considered the most powerful labor leader in the country after Meany, did not escape Johnson’s attention. “Never before had Reuther been on such close, continuous, and informal terms with an American President,” writes historian Nelson Lichtenstein (1995, 389). “Throughout 1964 and 1965 there were phone calls almost every week, dinner invitations to the White House, and the kind of open political collaboration that Reuther craved,” Lichtenstein recounts.

When, shortly before leaving office, President Johnson made a final presidential visit to AFL–CIO headquarters, he offered a quantitative measure of his interaction with President Meany: “I looked over my diary last night and I have met with Mr. Meany…forty nine times, in personal meetings either in my office, the Oval Room, or the Mansion. In addition to that, he has called me, or I have called him, eighty-two additional times” (Johnson 1969). The President then presented Meany with a framed set of 100 pens used in signing 100 major pieces of legislation during the Johnson presidency (a display which hangs in the AFL–CIO Department of Legislation to this day).

As Johnson’s gift suggests, labor’s presidential access was associated with remarkable political benefits. Johnson worked with union leaders and lobbyists to orchestrate the enactment of what George Meany called “a tremendous volume of long-overdue social welfare legislation” (AFL–CIO 1965). New laws in civil rights, health care, education funding, public housing, the minimum wage, and anti-poverty policy were all lauded by labor leaders. Less satisfying, though, was the outcome on the labor movement’s effort to repeal the “right-to-work” component (Section 14b) of the Taft–Hartley Act of 1947, which allowed states to ban mandatory “agency” fees in unionized workplaces (Gall 1988; Johnston 2015). While a repeal bill easily passed the House of Representatives in 1965, it was defeated by a filibuster led by Senate Republicans. Still, labor’s organizational interests were protected through pro-union appointments to the Department of Labor, the National Labor Relations Board (NLRB), and other labor-related agencies, all of which were made through a process of consultation with the Federation and relevant individual unions (Meany 1969; Goulden 1972, 346; Moe 1987). The White House “checked with the AFL–CIO before they did much of anything,” recalled a key Federation staffer (Minchin 2017, 26).

The arrival of Richard Nixon in the presidency in January 1969 might have meant, in contrast, the beginning of a period of severe labor exclusion. The conditions certainly were in place. In the 1968 presidential election, a united labor movement had enthusiastically rallied around the candidacy of Democratic nominee Hubert Humphrey, organizing a campaign effort that journalist Theodore White called “unprecedented in American history” for its scale and expense (White 1969, 453). Unions fiercely attacked Nixon for his long congressional record of voting against union-endorsed bills, and his role (as a freshman in Congress, no less) in helping to draft the reviled Taft–Hartley Act (Black 2007, 89–91; Goulden 1972, 404; Robinson 1981, 278). Despite all this, once in office the new Republican president, like Johnson before him, felt it wise to grant generous access to the labor leadership, and take their preferences into account in important decisions and appointments.

“We want to keep open and friendly communication, notwithstanding any political differences,” wrote Special Counsel Charles Colson in a memo to the president about how to attract union support (Minchin 2017, 40). Nixon evidently took Colson’s admonitions seriously: He invited Meany to visit the White House on a regular basis, played golf with him, and spoke on the phone frequently (Minchin 2017, 40; Goulden 1972, 404–406). The administration also routinely sent top officials, including cabinet secretaries, to AFL–CIO conferences and events, and invited the Federation’s Executive Council to the White House for meetings with the president (Robinson 1981, 282). The most extravagant initiative came on Labor Day in 1970. The President decided that the White House should host a formal dinner for the top union leadership, along with a gathering on the White House lawn for an additional 4000 union families (Cowie 2002; Robinson 1981, 298–299; Wehrle 2008). Goulden captures the splendor of the event: “Nixon staged an extraordinary festive evening… The affair had the trappings of a full state dinner, replete with an armed forces string band playing softly just off the East Room, seven courses of fine food, and three wines. Mr. and Mrs. Meany sat at the head table and chatted with the Nixons during dinner” (Goulden 1972, 417). The evening included effusive and friendly toasts between Nixon and Meany, the texts of which were later used by Republican campaign operatives (to Meany’s annoyance) to imply a burgeoning alliance between organized labor and Republicans (Goulden 1972, 419).

Behind Nixon’s maneuvers were larger strategic considerations. At the electoral level, the President was fascinated by the idea of luring northern white workers away from the Democratic Party, thus rendering a death blow to the New Deal coalition, and he reasoned that friendly ties with union leaders would be crucial in accomplishing this goal (Cowie 2002, 268; McCartin 2013, 137). In addition, he sought to reward and amplify union support for his foreign policy, especially the escalation of the Vietnam War. In a time of profound dissent and skepticism, the support of the hawkish George Meany and other union leaders was immensely valuable (Wehrle 2008). But Nixon’s efforts to bring labor unions into the Republican fold were also incentivized and facilitated by the relatively relaxed norms of partisanship prevailing during this period. Nixon could make pro-union executive branch appointments and even support union-sympathetic legislation without running afoul of a congressional Republican party that would attempt to block him—there were, after all, still many moderates and even liberals among the party’s congressional delegation (Rae 1989). Likewise, unions could ally with Republicans on particular issues, and perhaps even endorse Republican candidates, without bringing upon themselves punishment or exclusion by Democrats in Congress and elsewhere—the congressional Democrats were, it turned out, just as divided as the Republicans, with many members just as happy as labor to cut their own deals with the President.

As during the Johnson presidency, there were gains for labor that could be attributed to the persistence of an ethos of bipartisan bargaining and associated modes of union access. In terms of appointments, Nixon’s labor secretaries were all acceptable to the AFL–CIO, with George P. Schultz (the first of three during Nixon’s term) a personal favorite of Meany (they maintained regular contact and become longtime personal friends) (Goulden 1972, 406). In 1973, at the start of his second term, Nixon named a building trades leader, Peter Brennan, as his secretary of labor, just as Eisenhower had done. The politics of the National Labor Relations Board also remained enmeshed in a postwar equilibrium that encouraged the appointment of legal professionals committed to predictability and continuity in labor law (Moe 1987). Reflecting these norms, Nixon’s choices for the labor board were considered moderates who acted in good faith, and “a disappointment to business conservatives organizing to control the NLRB” (Cowie 2002, 258; Gross 1995, 223–231). Nixon even supported the enactment of a labor-endorsed bill that would permit striking construction workers in a single union to picket an entire worksite (despite the presence of workers in other unions, under different contracts) (Robinson 1981, 279). Although Congress did not approve the legislation, the president did sign two bills that helped public employees: one granting early retirement to air traffic controllers, and another that enhanced the organizing and bargaining rights of postal employees (McCartin 2013, 141–142; Chaifetz 1971). Nixon’s efforts to aid public sector workers so impressed Meany that he asserted that Nixon had “certainly been more liberal than any man in the White House before him because he has come out for collective bargaining for government employees at every level” (Robinson 1981, 294). And, famously, Nixon approved a host of liberal bills passed by the Democratic Congress, including the creation of the Occupational Safety and Health Administration (long sought by unions), the establishment of an earned income tax credit, and expansions of Social Security and the minimum wage.

Nixon’s efforts reached an unexpected and remarkable culmination in 1972, when the AFL–CIO Executive Council voted to remain neutral in the general election contest between Nixon and Senator George McGovern (although individual unions retained the prerogative to make endorsements if they wished, and several eventually supported the Democratic nominee). There were several reasons for the decision, which broke with decades of Federation support for Democratic presidential candidates. Meany and his key union allies disliked McGovern for his dovish positions on the Vietnam War, and his liberal positions on various cultural issues. Perhaps most important, though, was the unease evoked by the Senator’s innovative route to the presidential nomination, which involved bypassing the party’s customary power brokers, including organized labor, and engaging instead in a grassroots mobilization of anti-war activists and other liberals in the presidential primaries (Goulden 1982, 216; Wilson 1979, 48–49). At the 1972 nominating convention, many union leaders felt ignored, even excluded, by state delegations in which they had little influence (Shafer 1983, 97). In this context, the assiduous and even deferential courting by President Nixon of the labor leadership, especially of George Meany and the conservative building trades, was surely even more appreciated. Nonetheless, many unionists, especially those in the industrial unions and the rising public sector, continued to despise and distrust Nixon, and felt that the AFL–CIO’s neutrality in 1972 was foolhardy and a betrayal of labor’s best traditions (Goulden 1982, 215–219). Undoubtedly, the labor/Nixon relationship contained major limitations. All the same, it reveals the flexibility of alliances and ideologies during the age of bargaining, and the degree to which presidential access occurred regardless of partisan affiliation.

George W. Bush, Barack Obama, and the age of polarization

By the beginning of the twenty-first century, the type of bipartisan bargaining and labor outreach that Richard Nixon (and other Republican presidents, such as Dwight Eisenhower and Gerald Ford) had embraced was largely forgotten, and indeed hard to imagine. The parties now clashed with unrelieved ideological fervor, bringing profound consequences for union access. During the presidency of George W. Bush, unions found themselves excluded from the executive branch to a degree not experienced since the 1920s. Conversely, the Obama administration forged a level of inclusion that had much in common with that which unions had secured under Lyndon Johnson and other Democratic presidents. The starkness of the change between the Bush and Obama administrations, while shocking, would prove entirely characteristic of the extremes fostered by a polarized system.

“I think we were hurt during the Bush years, because during that period there was no communication whatsoever between the White House and organized labor,” recalled AFL–CIO Secretary-Treasurer Linda Chavez-Thompson (Minchin 2017, 268). Even the AFL–CIO President, John Sweeney, found himself completely frozen out: he reported that he was invited to the White House exactly once during George W. Bush’s 8 years in office, and then only to meet visiting Pope Benedict XVI, who had personally requested that Sweeney attend (Greenhouse 2009). The contrast with the norms of the age of bargaining was startling: instead of being feted at elegant White House dinners by a Republican president, the leader of the nation’s labor federation, still representing 13 million workers (in the early 2000s), was effectively excluded from the White House altogether. “White House officials have repeatedly said they see little reason to meet such an implacable foe,” the New York Times reported (Greenhouse 2003).

For the Bush administration, organized labor had simply become a group aligned with the opposing party—indeed, it was viewed as essentially a part of that party. White House Press Secretary Ari Fleischer declared that the Federation’s leaders were now, in a clear negation of Meany’s strategic dictum of the 1950s, little more than “appendages of the Democratic National Committee” (Minchin 2017, 267). In the prevailing political climate, no regeneration of union resources, however large, was likely to change that judgment, or incentivize the administration to cultivate the support of the AFL–CIO. And efforts in that direction were sure to be met with hostility from many Republicans on Capitol Hill and the party’s increasingly militant base (Roof 2011, 209). To be sure, the logic went both ways. For the unions, a better relationship with Bush would have likely brought dissent from within their own membership, and from Democratic allies in Congress and elsewhere (Lichtenstein 2011, 2015). In this sense, Fleischer’s judgment, while inflammatory, was not far from the truth. The union movement as a whole was clearly more liberal and partisan than it had been a generation before (Lichtenstein 2015). The growth of public employee unionism especially encouraged liberal activism as unionists sought to elect officeholders who, in addition to acknowledging their right to exist, were willing to accept more generous contracts (and allow the spending, and possible tax increases, they implied) (DiSalvo 2015; Siegel 2012). With few exceptions, such officeholders would be Democrats. Moreover, a union movement with a larger percentage of women, racial minorities, and educated professionals was already primed to move toward the political left (Camobreco and Barnello 2015). While the building trades and other moderate unions remained a powerful grouping, the election of John J. Sweeney as AFL–CIO President in 1995 reflected the emergence of public and service sector unions, and their liberal industrial union allies, as the leading force within the federation (Dark 1999).

The consequences for the access and influence of labor were clearly evident in George W. Bush’s approach to the Labor Department. With virtually no consultation with unions, he first nominated Linda Chavez, a former lobbyist for the American Federation of Teachers who had become a conservative columnist and strident critic of unions (later even penning a volume entitled Betrayal: How Union Bosses Shake Down Their Members and Corrupt American Politics) (Chavez 2004). Despite her union background, she opposed proposals to raise the minimum wage, and criticized affirmative action programs enforced by the department (Holmes and Greenhouse 2001). Angered by what they considered her betrayal of the labor movement, unions were aghast at Chavez’s selection. The AFL–CIO Executive Council officially opposed her confirmation, insisting that she be “rejected for her consistent and vitriolic opposition to the many laws and regulations that she, as secretary of labor, would be charged with upholding and enforcing.” (AFL–CIO 2001). Ultimately, Bush was forced to withdraw the nomination when it was revealed that Chavez had skirted wage and immigration laws in hiring a live-in maid. Nonetheless, the episode made clear where labor stood in the Bush White House under conditions of fully polarized politics. Bush’s second choice, confirmed quickly by the Republican Senate, turned out to be almost as disappointing. Elaine Chao was a former business executive, an analyst at the Heritage Foundation and, not insignificantly, the wife of conservative Kentucky Republican Senator Mitch McConnell. During her 8 years as labor secretary, unions criticized her for relaxing business regulations, rolling back federal safety standards, and harassing unions with excessive reporting requirements (Dean 2016). In 2003, Sweeney asserted: “I have served on the AFL–CIO Executive Council for 22 years. I have met and worked with every Secretary of Labor over those 22 years… I have never met a more anti-worker, anti-union Secretary of Labor than Elaine Chao. They clearly are trying to go after us in every way they can” (Lane 2003).

Unions complained vociferously about a series of Bush administration decisions that, in their view, were unusually hostile to union interests. Only 2 months into the new administration, the President signed legislation repealing union-backed ergonomics regulations that had been adopted at the close of the Clinton administration (Roof 2011, 207). The President also issued executive orders limiting the use of union member dues for political activities and ending preferences for unionized companies in federally financed building projects (Greenhouse 2001). Bush strongly opposed collective bargaining rights for employees at the new Homeland Security Department (in stark contrast to Nixon’s more conciliatory attitude toward federal employee unionism) (Roof 2011, 208). Perhaps most dramatically, the mid-century norms of professionalism that once guided NLRB appointments were now replaced by incessant ideological warfare over appointments, the increased use of recess appointments, and long periods during which board seats were left vacant (thus precluding the legally mandated quorum for board decisions) (Moe 1987; Gould 2015). In an analysis of these developments, labor relations specialist Gerald Friedman concluded: “Bush came to office without ties with labor and used every opportunity to challenge labor’s legitimacy and to attack unions” (Friedman 2010). The labor movement as a whole was simply no longer viewed as even a potentially useful political partner by the Republican White House.

In the context of mature polarization, the arrival of a new Democratic president would be expected to bring an immediate and sharp reversal in union access. And so it did. A few months into the presidency of Barack Obama, Sweeney reported that he was now visiting the White House at least once a week (Greenhouse 2003). His successor, Richard Trumka (elected to replace the retiring Sweeney as Federation President in September 2009), was treated similarly. In comments that recalled almost word to word Meany’s boastful description of his access to LBJ, Trumka remarked in 2011: “I’m at the White House a couple of times a week—two, three times a week. I have conversations every day with someone in the White House or the administration—every day” (O’Brien 2011). Service Employees International Union (SEIU) President Andy Stern, widely regarded as the most influential union leader after Trumka, observed: “We get heard. SEIU is in the field, it’s in the White House, it’s in the administration” (Maher 2009). The comments by Trumka and Stern were confirmed by the official records of White House visitors provided by the Obama administration. The list included many officials from the AFL–CIO’s own bureaucracy, individual national unions, and the smaller Change to Win federation (a breakaway group of unions that had formed in 2005). The total number of visits by union officials were the most of any interest groups listed, exceeding those of other Democratic-leaning groups as well as business leaders (Weisman and McKinnon 2009).

The level of access to the president by SEIU’s Andy Stern was so profound that The Nation’s Max Fraser wrote that Stern “became one of Obama’s closest confidants outside of government—a throwback to the days when union presidents doubled as labor statesmen on the national stage” (Fraser 2010). As the administration turned to the issue of health-care reform, Stern and SEIU staff worked closely with the White House and business interests to help broker agreements conducive to legislative success, much as the AFL–CIO had helped pass Medicare and other social welfare bills in the 1960s (Greenstone 1977; Roof 2011). Stern’s distinctive place as a leader with access to the president, key players in Congress, union leaders, and top business figures made him an unusually valuable figure in the health-care struggle (Daschle 2010, 83–87; Alter 2010, 251–252; Early 2010). “I think there was a role for someone in labor to provide the unique contribution that Andy has,” former Democratic Senate Majority Leader Tom Daschle said. “You needed somebody to take some risks and step up to the plate—it was ready-made for a labor leader” (Lehman 2010). While this integrative role was described by The Nation as a “throwback,” it should not be seen as simply an antediluvian legacy of past ties; it grew quite naturally out of a system of polarized partisanship in which party/interest group linkages are reinforced and deepened.

The presence of a liberal Democrat in the White House, and one prepared to grant labor regular access, itself reflected the effects of party realignment. The previous three Democratic presidents had been southern white men with affiliations, to varying degrees, with the politics of the more conservative Democratic Party native to their region. Lyndon Johnson, Jimmy Carter, and Bill Clinton had each begun their political careers in southern right-to-work states with small labor movements. Carter and Clinton had also expressed a desire to modernize the political orientation of the Democratic Party, an ambition that often implied a downgrade in the status of organized labor. Barack Obama brought something different to the table. As an African–American northern liberal Democrat, he not only depended on union support in his earlier campaigns for office, but had worked with unions as a community organizer in Chicago and as a state legislator (Garrow 2017, 795–855; Remnick 2010, 370). In The Audacity of Hope, Obama unabashedly celebrated his ties with unions: “For 7 years I had been their ally in the state legislature, sponsoring many of their bills and making their case on the floor.” Noting the endorsements of his 2004 Senate campaign by SEIU, the American Federation of State, County, and Municipal Employees (AFSCME), UNITE HERE, and the Illinois Federation of Teachers, Obama wrote: “So I owe these unions. When their leaders call, I do my best to call them back right away. I don’t consider this corrupting in any way; I don’t mind feeling obligated toward home health-care workers who clean bedpans every day for little more than minimum wage” (Obama 2008, 142). This recognition of union influence and the legitimacy of union access could have just as easily been uttered by Democratic politicians at the highpoint of the age of bargaining.

Once in office, Obama continued to evoke pro-union themes. In his 2015 State of the Union address, he told Congress: “We still need laws that strengthen rather than weaken unions, and give American workers a voice” (Obama 2015). Craig Becker, who served as the AFL–CIO’s legal counsel and also an Obama appointee to the NLRB, wrote in 2015: “President Obama has spoken more openly about the importance of unions than any president since Franklin Roosevelt—and Obama’s advocacy of union membership, unlike Roosevelt’s, extends to the public sector” (Becker 2015, 65–68). Similarly laudatory claims were made by the AFL–CIO’s Legislative Director, William Samuel: “I actually think this president and this White House has a much better feel for the labor movement than the Clinton White House. They [the Clinton administration] would be critical of the labor movement because they were New Democrats. That’s not what this White House is about… I think there’s a much more organic understanding of the labor movement and our importance” (McMorris-Santoro 2013; Samuel 2015).

Of course, President Obama’s “understanding” of the labor movement was no guarantee that he would give first priority to union interests—as, indeed, he did not (just as Lyndon Johnson had not, nor other Democratic presidents before him) (Johnston 2015; Gall 1988). When it came to labor law reform, specifically the passage of the Employee Free Choice Act (EFCA) establishing a “card-check” method of union recognition, Obama followed in the footsteps of LBJ and gave health care, the stimulus bill, and other measures precedence. The unfortunate result for labor was that political momentum for EFCA faded and, with a Senate filibuster looming, the bill never made it to the floor in either chamber of Congress (Warren 2011). The failure once again to obtain a revised legal framework to better protect unions was a conspicuous example of the limits of labor power even when political access was well established. But access alone could never be expected to provide a means to compel Senate Democrats from states with low union density to join a Senate cloture vote—a vital prerequisite for passing a controversial pro-labor reform bill (Roof 2011, 224–226). For that, a revival of labor’s organizing fortunes or disruptive capacity would almost certainly be required—a task that seemed well-beyond the capability of the existing labor movement (Friedman 2009; Rosenfeld 2014; Lichtenstein 2002).

Despite the failure of EFCA, Obama made appointments to the Department of Labor and the NLRB that unions considered excellent, and issued an array of executive orders that advanced union interests (Noah and Mahoney 2015; Baker 2015; Lee 2015). His second labor secretary, Thomas Perez, was lauded as the “best labor secretary ever” by the AFL–CIO’s Director of Governmental Affairs (Samuel 2015). CNN reported: “National union leaders say they have not seen the kind of popularity Perez enjoys among their members in years” (Lee 2015). At the NLRB, Obama ultimately succeeded in shifting the board in a pro-labor direction, making a series of successful nominations following the 2013 abolition of the Senate filibuster on presidential appointments. “The quality of this board is the best ever,” said Larry Cohen, president of the Communications Workers of America, in 2015. “The NLRB appointments is one place where President Obama would get a perfect score” (Noah and Mahoney 2015).

All in all, these were familiar patterns and familiar relationships. To be sure, organized labor had to share access with many more interest groups and social movements than existed in the 1960s (Zelizer 2004, 263), and its overall centrality to American politics and economics was undoubtedly diminished. Still, it would be very difficult to say that the nature of union access had fundamentally changed during the Obama presidency (much less collapsed, as some media and political narratives would have it). As always, access meant only an opportunity to influence presidential priorities and strategies. Inevitably, unions would emerge from their interactions disappointed on a host of issues—but such tensions were hardly a new feature of the relationship. Even Lyndon Johnson, after all, had been attacked by labor leaders for failing to deliver repeal of Section 14b of the Taft–Hartley Act and endorsing union-opposed wage and price guidelines (Goulden 1972, 350–351).

Union resources and the case of Donald J. Trump

In contrast to the above account, a resource-based viewpoint offers a different mode of explanation, one that is often presented as sufficient in itself to explain changes in levels of access. This view takes union density as a definitive indicator of overall union resources, and on this basis posits a sharp decline in political power and access across the board, including within the Democratic Party (Andrias 2016; Rosenfeld 2014). As one recent analysis concludes: “The most obvious reason for the diminished political influence of labor is that, as union membership has plummeted, unions have had fewer workers to mobilize in politics and fewer resources to deploy on behalf of workers’ goals” (Andrias 2016). However, the assumption that declining membership equals declining resources is not obvious—it is always possible that a group will compensate for membership decline by extracting resources more effectively from a smaller base, or by finding alternative sources of support (investments, grants, etc.), or by simply adopting more efficient techniques for managing existing resources. Indeed, there is an abundance of evidence that unions have actually done all of these things, and that this has enabled them to remain a powerful force in campaign finance, electioneering, and lobbying, even vis-à-vis their business opponents (Camobreco and Barnello 2015; DiSalvo 2015; Francia 2006, 2012; McGinty and Mullins 2012; Minchin 2017; Roof 2011; Siegel 2012).

The sheer size and scope of union resources should not be underestimated. Although density has declined, the labor movement retains a substantial dues-paying membership that enables impressive electoral mobilization and massive fundraising. In 2017, the AFL–CIO still had 12.5 million members, and there were another 2.3 million members in unions unaffiliated with the Federation. Out of that 14.8 million total, 7.2 million were in the public sector, and 7.6 million in the private sector (Bureau of Labor Statistics 2018). These numbers allowed unions to continue to comprise a significant percentage of the total electorate. According to the Roper Center for Public Opinion Research, voters from union households were 26% of the national electorate in the 2000 presidential election, 24% in 2004, 21% in 2008, and 18% in both 2012 and 2016 (Roper Center 2018). Voters from union households were an even larger percentage in crucial presidential swing states, composing 28% in Michigan and 21% in Wisconsin in the 2016 election (CNN 2018). Beyond these percentages, numerous studies show that union campaign activity effectively increases the turnout of union members (and their families)—especially among low to middle-income whites, the less educated, and minorities—and skews those votes in a Democratic direction (Francia 2006, 2012; Rosenfeld 2014, 170–172; Enten 2014; Silver 2011).

Union spending on elections has also remained impressive through the 2000s. For the 2015–2016 election cycle, one analysis (based in part on required union reports to the US Department of Labor) concludes that unions spent a total of at least $1.7 billion at all levels of American politics (National Institute for Labor Relations Research 2017). This total includes $1.3 billion drawn directly from union treasuries (as expedited by the Supreme Court’s 2010 Citizens United v. FEC decision) for various forms of political education and mobilization (including donations to other liberal advocacy groups). In addition, union political action committees contributed $58 million directly to federal candidates in the 2016 cycle, and spent $149 million on soft or outside money contributions; another $9 million was contributed directly by individuals affiliated with labor unions (Center for Responsive Politics 2017). State and local disclosure reports (which include information from state-level public employee unions who do not report to the federal Department of Labor) show at least another $228 million spent by unions (National Institute for Labor Relations Research 2017). Since there is good reason to believe that union political expenditures are under-reported (McGinty and Mullins 2012), it is entirely possible that the total amount of union money deployed in American politics in 2015–16 approached $2 billion. While the business sector clearly spent much more in total (for example, at least $3.3 billion in federal-level contributions alone in 2016), the sum spent by labor would certainly concentrate the mind of any election-focused politician (Center for Responsive Politics 2017). And the approximately $45 million that unions have spent on federal-level lobbying each year for the last 10 years is surely of some import on Capitol Hill (Center for Responsive Politics 2017).

Even after decline, then, it is clear that unions have continued to generate large amounts of politically valuable resources. However, in recent decades Republican officeholders, including presidential candidates, have rarely bid for such resources, or even made a serious effort to neutralize them. One reason is that the labor movement itself is more liberal, reflecting the growth of public employee unionism, internal demographic changes, and the rise to high union office of left-wing activists who first entered the labor movement in the 1970s (Lichtenstein 2015; DiSalvo 2015; Siegel 2012). The bulk of union votes and financial contributions continue to flow to Democratic candidates (88% of union donations to federal campaigns went to Democrats in the 2016 cycle) (Francia 2012; Rosenfeld 2014; Center for Responsive Politics 2017). Despite this, it remains the case that the well-funded building trades unions, who have declined more slowly than their industrial unions counterparts, are powerful and independent actors who remain open to bipartisanship. Moreover, unions of law enforcement personnel, firefighters, and other skilled trades (as well as the notorious Teamsters union) are also available to work with Republicans. Yet, the leaders of such unions were never even invited to visit the George W. Bush White House, and their needs have typically been ignored by contemporary congressional Republicans. The other side of the equation, then, has been decreasing Republican willingness to reach out to even those parts of the labor movement amenable to possible recruitment or mutual accommodation.

The partisan differences experienced by unions are, of course, hardly unique to labor organizations. As Sinclair noted over a decade ago, most interest groups have experienced increased pressure to decisively choose one of the two parties as their primary entry way into the political system. Sinclair (2006, 312) reports, for example, that Republican positions on property rights and environmental regulation have driven resource-rich environmental groups almost entirely into the Democratic Party, despite these groups’ historically bipartisan orientation. Conversely, Democratic support, however tepid, for gun control has sent the National Rifle Association almost entirely into the Republican camp, again in contradistinction to previous bipartisan alliances. Variations in resource levels are not the key issue in such developments; rather, ideologically coherent parties simply refuse to offer much to groups if those groups are too far from their own (increasingly strident) ideological position (which is itself buttressed by groups within their own coalition) (Karol 2015). In these circumstances, “party politics subsumes interest group politics; access is partisan” (Hansen 1991, 223).

Based on this analysis, we would expect that union access to Republican presidents would only change if the ideological orientation of the party—or at least its presidential nominee—underwent a major alteration. It is party politics itself that must change first, not the strategy or resources of the unions themselves. Prior to the election of 2016, such a transformation seemed thoroughly unlikely, for it would require the nomination of a Republican presidential candidate with deviating views on the relationship of the party to unions, and the general place of organized labor in American society. The nomination of a candidate who would seriously envision a return to a mid-century Eisenhower/Nixon approach to unions would have been considered a virtual impossibility. Yet, to a surprising (albeit limited) degree, Donald Trump would prove to be such a candidate. As his campaign for the Republican nomination unfolded, he chose to cast himself as a populist defender of the “forgotten” working man, and said he wanted to turn the Republican Party into a “worker’s party.” He envisioned “a party of people that haven’t had a real wage increase in 18 years, that are angry…” (Green 2016). As part of this strategy, he chose to emphasize his positive relationships with unions as a New York real estate developer and the owner of unionized casinos and hotels. “I have great relationships with unions,” he said. “New York is mostly unionized” (Cooper 2015). While he criticized the AFL–CIO for endorsing Hillary Clinton, and offered no proposals for an enhancement of union organizing rights, he also never attacked the legitimacy of unions as such, and certainly not the institution of collective bargaining. And Trump clearly aimed to gain the support of union members, especially in the swing states of the industrial Midwest. Union leaders, in turn, scrambled to try to prevent their members from succumbing to Trump’s appeals.

On Election Day, Trump emerged with 43% of the union household vote to Clinton’s 51%—the narrowest margin since Reagan’s 1984 landslide (Bump 2016). Among whites from union households with no college degree, Trump actually won by a large margin: 58–32% (Meyerson 2017). With his repeated references to the “silent majority” and his open contempt for liberal media and urban elites, Trump’s discourse bore a marked resemblance of that of Richard Nixon in his two successful presidential campaigns. And, as with Nixon, his strategy suggested a plan to accentuate the divide between more conservative unions, whose members are more likely to be composed of white men in skilled trades, and the most liberal unions, whose members are more likely to be female and/or non-white and located in the public sector and related service industries (Meyerson 2017; Yeselson 2017).

After his victory, Trump might have chosen to continue to appeal directly to workers, including union members, over the heads of the union leadership. Instead, and as if to underscore his difference from recent Republican presidents, Trump summoned AFL–CIO President Trumka to Trump Tower on January 13, 2017 for a 90-min meeting. Trumka told reporters that the summit was “very productive” and that they spoke about “a lot of issues” (Kamisar 2017). Also during the transition, Trump met with Teamsters President James Hoffa and James Callahan, President of the International Union of Operating Engineers (Karni 2017; Shepardson 2017). Once installed in the White House, Trump continued to convene regularly with unionists. Just 3 days into his presidency, the President arranged a highly publicized meeting with six union presidents (and other union activists) from the construction industry, who were joined in the Oval Office by the Vice President and other top officials (Scheiber 2017). In April 2017, he met again with Teamsters President Hoffa and Vice President John Murphy to discuss pension policy and “other issues” (Teamsters 2017). Trump continued to meet with Trumka during the first year, discussing “wages, trade and infrastructure” issues, according to press reports (Merica 2017). In February 2018, Trump even met with the leaders of the more liberal industrial unions, bringing the presidents of the United Autoworkers, the International Association of Machinists, the United Steelworkers, the Communication Workers of America, and the Teamsters (as well as President Trumka) to the White House to discuss the terms for renegotiating NAFTA (the North American Free Trade Agreement) (Higgins 2018).

Trump’s willingness to provide privileged access to at least parts of the union leadership was reflected in some substantive policy shifts. Trump withdrew US support for the Trans Pacific Partnership trade agreement, began the process of NAFTA renegotiation (with union consultation), and adopted new tariffs on steel and aluminum imports (supported by industrial unions). He approved the Keystone XL Pipeline and Dakota Access Pipeline, both of which would provide jobs for unionized construction workers, and he pledged that he would pursue major increases in infrastructure spending. He also pulled out of the Paris climate accords—a decision that could help protect unionized coal workers. Notably, these measures mainly benefitted construction unions and, to some degree, the industrial unions that still held considerable influence in Midwestern swing states. As for the public employee and service unions that President Obama had so strongly favored in White House visits and consultations, Trump offered virtually nothing.

Notwithstanding the quality of access, the limits of Trump’s rapprochement with labor became evident over the course of his first year in office. At the Labor Department, Trump first nominated as secretary a conservative fast-food executive, Andrew Puzder, who had been critical of minimum wage and overtime laws, and who was accused by the AFL–CIO of routinely violating labor law. Puzder eventually withdrew from consideration after widely publicized accusations of spousal abuse and his hiring of an undocumented maid. Trump then nominated Andrew Acosta, a law school dean and former member of the NLRB, who was universally considered a more moderate choice, and whose confirmation was not opposed by the labor movement. In the case of the NRLB, unions would not be so lucky, as Trump made two new appointments to the board, as well as a new general counsel, that shifted the NLRB in a much more conservative direction (Vail 2018). Once again, the board continued to veer sharply in response to partisan appointments (notwithstanding earlier postwar norms and expectations) (Moe 1987). Unions also complained when Trump supported the rollback of union-supported regulations issued during the Obama presidency, and signed legislation overturning a rule requiring federal contractors to disclose wage, safety, and workplace discrimination violations. Evaluating Trump’s first year in office, Trumka was highly critical: “Broken promises are bad enough. But President Trump has also used his office to actively hurt working people. He has joined with corporations and their political allies to undermine the right of workers to bargain collectively. He has taken money out of our pockets and made our workplace less safe. He has divided our country, abandoned our values and given cover to racism and other forms of bigotry” (Tackett 2018). Still, even after this condemnation, Trumka was invited to the White House along with industrial unions to discuss trade policy; he continued to have a level of access that was never achieved by AFL–CIO leaders in 8 years of the George W. Bush presidency.

One year into his term, Trump had demonstrated serious interest in gaining the support of selected unions, both through privileged access to the Oval Office and by providing a set of particularistic benefits—mainly building projects and trade protection. Trump did not, however, show any willingness or interest in going much further than this, and his fellow Republicans on Capitol Hill expressed even less enthusiasm. Even so, Trump had partially broken with the logic of contemporary partisan polarization, and done so in a manner that had authentic policy content. In the 2016 election and as president, Trump has been sufficiently unencumbered by ideological orthodoxy to actively seek the support of the white working class, union members, and even union leaders—a populist “blue-collar” option that had long been on the table, if unappealing to many party leaders (Miller and Schofield 2008). To the surprise of most Republican strategists, Trump’s approach proved electorally successful in 2016, and raised the specter of a further recasting of party alignments, one that could potentially lead the Republican Party in the direction of becoming the party of trade protection, immigration restriction, and, perhaps, a considerable portion of private sector unionism.

Conclusion

As in other areas of American politics, polarization has brought a new dynamic to the relationship between interest groups and elected officials, establishing partisan affiliation as a dominant element in political calculation and alliance formation. In recent decades, the shifting of party coalitions and associated purification of party ideologies brought unions and Democrats closer together, while pushing unions and Republicans further apart. Consequently, the partisan affiliation of the president took on much greater importance as an independent determinant of union access. Unions found themselves whipsawed between continued access during Democratic administrations, and exclusion during Republican governance. In this sense, polarization mitigated the effects of decline on the power of labor during Democratic presidencies, while exacerbating it under Republican rule. Although union decline—in both economics and politics—has certainly occurred, its effects have been refracted through the prism of a polarized political system. Polarization thus brings its own causal force, distinct from raw metrics of interest group power, to the union/president relationship, and to interest group politics more generally.

Of course, these relationships are not immune to change. If a president shifts party ideology in a new direction, essentially depolarizing the parties in a particular issue area, interest groups previously embedded in a single party coalition may become available again for cross-partisan alliance formation and bargaining. Such innovations are inherently difficult, however, as they threaten existing party ideologies and interests, which tend to become deeply entrenched in a polarized context (Miller and Schofield 2008). Donald Trump’s efforts to recruit union support were limited by the hostility to organized labor common among congressional Republicans and in Republican-dominated statehouses. While Trump did have his own political base, as a practical matter he was highly dependent on traditional Republicans for legislative support and administrative personnel. Thus, his attempt to at least partially depolarize labor policy and bring private sector unions into a recast Republican Party was a truly herculean endeavor (and one which he showed a limited capacity to effectively pursue). While it was impossible to predict where Trump’s unorthodox initiatives in labor access might lead, his departures from the norms of polarized politics appeared unlikely to be sustained in the absence of a larger transformation of both Republican and Democratic party coalitions.