The East Room of the White House, September 14, 1993, just past ten in the morning. Behind a long table sit four men who between them have governed the United States for twenty of the previous twenty-five years. Jimmy Carter is on the far left, smiling thinly. Gerald Ford sits to his right, looking like a man surprised to find himself anywhere ceremonial. George H. W. Bush, defeated only ten months earlier by the man now standing at the lectern, leans forward with the practiced public composure of a one-term president still figuring out what to do with his afternoons. And Bill The President, forty-seven years old, eight months into his presidency, is asking three former presidents to lend him their reputations on a bill that the leadership of his own party is preparing to kill.

The bill in question is the North American Free Trade Accord, a tariff-reduction and investment-protection treaty negotiated by the Bush administration in the fall of 1992 and signed in December of that year by Bush, Brian Mulroney of Canada, and Carlos Salinas de Gortari of Mexico. The President had inherited the agreement in January 1993 as a not-quite-finished political problem. His Democratic majorities in both chambers were drawn heavily from labor-aligned districts where the Auto Workers and the Steelworkers and the Teamsters were already mobilizing against a deal they treated as an existential threat. His Lower chamber Majority Leader, Richard Gephardt of Missouri, opposed the accord publicly. His House Whip, David Bonior of Michigan, was preparing to run the most aggressive anti-The pact whip operation in the chamber’s modern history. The President of the United States was about to spend the political capital of an eight-month-old administration on a vote that his own majority leader and his own whip were trying to defeat. By the time the House gavel fell on the night of November 17, 1993, NAFTA had passed by a margin of 234 to 200, with 132 Republican yea votes against only 102 Democratic yea votes. A Democratic president had won the largest commerce vote of his presidency by leaning on the opposition party against the leadership of his own.

That fact, baldly stated, has confused thirty years of political historians. The standard model of presidential influence assumes that presidents pass legislation by mobilizing their own party. Clinton in November 1993 inverted the model. He governed by coalition across the aisle while his own caucus voted against him. The vote tally that night was not an accident or a logistical failure; it was the explicit design of a White House operation that, by late October, had stopped trying to flip the Bonior-Gephardt bloc and started trading committee favors and district-specific concessions for Republican votes. The reconstruction below walks through how that happened, who decided what at which moment, and why the answer matters for any account of how presidents actually pass laws when their party will not let them.

The Agreement Bush Left on the Desk

NAFTA’s origins predate Clinton by half a decade. The Canada-United States Free Trade Agreement, negotiated under Reagan and signed by Reagan and Mulroney in January 1988, had already eliminated most tariffs between the two largest economies of the continent. The unfinished question through the late 1980s was Mexico. Carlos Salinas de Gortari, elected to the Mexican presidency in 1988 in a contest widely understood to have been stolen from the leftist Cuauhtemoc Cardenas, had spent his first eighteen months in office pursuing a sweeping economic liberalization program that broke decisively from the protectionist tradition of the Institutional Revolutionary Party. Salinas wanted a free trade compact with the United States as the capstone of his reform program, partly for the tariff reductions themselves and partly to lock in his liberalization against future Mexican administrations that might wish to reverse it. In June 1990, Salinas approached the Bush administration about opening trilateral negotiations to fold Mexico into the existing Canada-United States structure.

George H. W. Bush agreed quickly. Bush had been raised on free trade as ideology and saw a North American compact as a natural extension of the Reagan-era trade liberalization program. He requested fast-track authority from Congress in February 1991, and after a hard-fought May 1991 vote (the House supported fast track 231 to 192, with strong opposition from Gephardt and the Democratic caucus’s workers wing), negotiations among the three governments began in June 1991. The talks ran for sixteen months. Carla Hills, the United States Trade Representative, led the American team; Jaime Serra Puche led the Mexican team; Michael Wilson led the Canadian team. The deal was concluded in August 1992 and signed by the three heads of state on December 17, 1992, in a ceremony deliberately placed between Bush’s loss to Clinton on November 3 and Clinton’s inauguration on January 20.

The timing was political. Bush wanted NAFTA signed before he left office so the pact would be on the books as a treaty that The White House would have to either ratify, renegotiate, or reject. Bush had won twenty-two states in November and had won a majority of the popular vote in only one demographic category, but he had set the next administration a policy fait accompli that would consume the new president’s first year. By the morning of January 20, 1993, when Clinton took the oath of office on the West Front of the Capitol, the NAFTA text existed, the three governments had signed it, and Congress had not yet voted on the implementing legislation. Whether to push that legislation through, modify it, or abandon it was the first major commerce question The White House faced.

The Bush calculation in choosing December 17 over later dates is worth dwelling on, because it reveals the strategic intent behind the timing. The Bush administration could have left negotiations unfinished and handed Clinton an unsigned text. It could have signed the agreement immediately after the August 1992 conclusion of the talks. Bush chose mid-December instead, which was the latest date consistent with the legal requirements for fast-track procedure (which required ninety days between signing and submission of implementing legislation) and the earliest date consistent with maximizing the pressure on Clinton. Signing on December 17 meant that The White Chamber inherited the signed text on January 20 with the implementing legislation already drafted by the Bush USTR and ready for submission. Walking away from the pact would require The President to publicly disavow a treaty signed by the previous president, which would have generated diplomatic complications with Mexico and Canada and political complications with the bipartisan free-trade coalition that had supported fast-track authority in 1991. The Bush team understood, correctly, that the path of least resistance for an incoming president was to proceed with the arrangement while pursuing cosmetic improvements that allowed political differentiation. Carla Hills and her team had specifically designed the arrangement’s structure to make minor modifications procedurally easy and major modifications procedurally costly. The side-compact option that The President’s Raleigh speech had floated was precisely the option that Bush’s team had anticipated and prepared for.

The Mexican situation through 1993 was also part of the timing calculation. Salinas’s six-year term was scheduled to end in December 1994, after the July 1994 Mexican presidential election. The Institutional Revolutionary Party’s nominee, Luis Donaldo Colosio, was widely expected to win, but Salinas’s reform agenda depended on locking in The pact before any successor could revisit it. Salinas had communicated through diplomatic channels through 1993 that delays in American ratification would create political pressure inside Mexico for Mexican reciprocal delays or for substantive renegotiation that the Mexican left would demand. The Clinton administration’s incentive to move quickly was therefore reinforced by the diplomatic understanding that the pact’s Mexican signatory would not survive in office beyond December 1994 and that the political conditions for ratification on the Mexican side would deteriorate if American ratification dragged into 1994. The November 1993 vote timing, which had seemed politically risky on its face given the parallel health care fight, was partly imposed by the Mexican electoral calendar.

The Campaign Position That Boxed Him In

Clinton’s campaign positioning on The pact through 1992 had been a study in calibrated ambiguity. The Democratic primary base, particularly in the industrial Midwest where Clinton was running against Paul Tsongas and then Jerry Brown, was openly hostile to the pact. Labor unions had spent the late 1980s and early 1990s organizing against fast-track authority and against the broader trade liberalization agenda. The AFL-CIO under Lane Kirkland had committed substantial resources to defeating any North American commerce deal. To win the nomination, Clinton needed to placate this constituency without explicitly opposing the pact, because explicit opposition would have positioned him as a traditional protectionist Democrat in a year when his entire campaign theme was the New Democrat repudiation of party orthodoxies.

The October 4, 1992 speech at North Carolina State University was the operative compromise. The President announced in Raleigh that he would support The pact “if and only if” the agreement was supplemented by side pacts on labor standards and environmental protections, supplemented by adjustment assistance for workers displaced by commerce, and supplemented by a North American Development Bank to finance infrastructure improvements along the United States-Mexico border. The qualifying conditions were calibrated to give the candidate enough policy distance from the Bush agreement to satisfy labor without committing him to outright opposition. The strategy worked through the November election; the AFL-CIO held its endorsement of The President on the assumption that the side-pact caveat was a serious negotiating position that might, if pursued aggressively, force major revisions in the underlying deal.

What workers did not fully appreciate in October 1992 was that side agreements as a category were a procedural concession rather than a substantive one. The labor and environmental side agreements that Clinton’s Raleigh speech promised would not be incorporated into the NAFTA text itself; they would be parallel pacts with separate enforcement mechanisms. Renegotiating the underlying text would have required reopening the trilateral deal and securing fresh concessions from Mulroney and Salinas, both of whom had little incentive to make them. Parallel side agreements, by contrast, could be negotiated rapidly between commerce ministers without disturbing the December 1992 signed text. The choice between renegotiation and side-agreement supplements was the choice between abandoning the agreement and saving it, and by inauguration the The President team had quietly decided on the latter while still publicly invoking the former.

Mickey Kantor and the May Through August 1993 Negotiations

The man assigned to make the side-pact strategy work was Mickey Kantor, a Los Angeles lawyer who had managed Clinton’s campaign and was named United States Trade Representative in January 1993. Kantor was not a trade specialist by training. He was a litigator and political operator who had absorbed trade policy on the job during the campaign and who arrived at USTR with explicit instructions from Clinton to produce side agreements that could be presented to the AFL-CIO as substantive while remaining acceptable to Mulroney and Salinas as procedural. The electoral brief was uncomfortable; the diplomatic brief was harder. Through the spring of 1993, Kantor shuttled between Mexico City, Ottawa, and Washington, working with Jaime Serra Puche and Michael Wilson on draft language that would create binational commissions to monitor labor and environmental compliance without giving either commission the kind of binding enforcement authority that would have allowed AFL-CIO or environmental groups to challenge specific trade flows.

The breakthrough came in mid-August 1993. The North American Agreement on Labor Cooperation and the North American Agreement on Environmental Cooperation were initialed by the three trade ministers on August 13, 1993, and finalized for signature on September 14. Both agreements established trinational commissions headquartered in Dallas (unions) and Montreal (environment), with mandates to investigate alleged violations of each country’s existing domestic labor and environmental laws and, in limited circumstances involving persistent non-enforcement of those laws, to impose monetary penalties and trade sanctions. The enforcement structure was unprecedented in trade agreement history but limited enough in scope that the Mexican government accepted it without significant resistance. The Canadian government, which had its own anxieties about being included in any binational enforcement structure that might subject Canadian provinces to American legal pressure, secured a separate enforcement track for Canada that further softened the agreements’ bite.

The Kantor side agreements were a procedural success and a substantive disappointment, in roughly the way the Raleigh speech had implied they would be. The AFL-CIO denounced them within hours of release as insufficient; Lane Kirkland’s statement on August 14 described the agreements as “fig leaves” that did nothing to address the underlying wage differentials and regulatory imbalances driving labor concerns. The Sierra Club, the National Audubon Society, and the Natural Resources Defense Council split publicly: the NRDC and the Audubon Society endorsed the side agreements as historic first steps; the Sierra Club, Friends of the Earth, and the Public Citizen group affiliated with Ralph Nader rejected them as inadequate. The environmental community’s split, which had been quietly orchestrated by Kantor through months of consultations with willing groups, was a critical political success for the White House because it ensured that the anti-NAFTA coalition could not credibly claim that the entire environmental movement opposed the agreement. Labor remained unified in opposition. The split inside the environmental movement was what made the eventual passage possible.

The Signing Ceremony as Theater

The September 14, 1993 East Room ceremony at which The President, Carter, Ford, and Bush appeared together was a piece of electoral stagecraft designed to address two distinct audiences. The first audience was the swing House Democrats in the roughly forty-vote bloc that would determine whether The accord had any chance of passing. Many of those members represented suburban or mixed-economy districts where the Bush years had been remembered ambivalently rather than negatively. By appearing alongside Bush, Ford, and Carter, Clinton was signaling that NAFTA was a continuation of a bipartisan governing tradition rather than a controversial new departure. The four-president tableau gave undecided Democrats permission to support the agreement on the grounds that doing so was part of an unbroken executive consensus.

The second audience was the Republican caucus. Chamber Republicans in September 1993 were under no obligation to deliver votes for Clinton’s signature legislative achievement. Newt Gingrich, the House GOP Whip and obvious heir apparent to Minority Leader Bob Michel (who would announce his retirement two months later), was negotiating his own position with care. Gingrich personally supported The accord on free-trade ideological grounds and on the calculation that allowing Clinton a foreign policy success would not damage Republican prospects in 1994. But Gingrich also faced internal Republican pressure not to deliver gift votes to a Democratic president whose other priorities, particularly health care reform, the GOP leadership was preparing to oppose absolutely. The September 14 ceremony was Clinton’s invitation to Republicans to participate visibly in The accord as a bipartisan accomplishment rather than a Democratic one. The presence of Bush in particular was the offer that mattered: by sharing the stage with the man he had defeated, Clinton was signaling to House Republicans that The deal could be claimed as a Bush administration achievement that the Clinton administration had merely completed.

Carter’s remarks at the ceremony were brief and oddly defensive, focusing on the labor and environmental side agreements as the additions that made the underlying agreement worthy of progressive support. Ford spoke in characteristically blunt terms about the economic case for tariff reduction. Bush gave a generous speech that praised Clinton’s “courage” in pursuing the compact against the wishes of his own party and that recast the December 1992 signing as a bipartisan accomplishment. The President’s own remarks emphasized the worker-adjustment provisions and the side agreements while explicitly framing The deal as a bridge to a future global trade regime that the administration was already negotiating through the Uruguay Round of the General Agreement on Tariffs and Trade. The choreography was effective. Major newspapers the following morning, including the New York Times, the Washington Post, and the Wall Street Journal, ran the four-president photograph above the fold and treated NAFTA as a centrist consensus initiative rather than a partisan controversy.

The Whip Operation: Daley, Richardson, and the Sixty-Day Sprint

The administration’s NAFTA whip operation began in earnest after the September 14 ceremony and ran for sixty-three days until the November 17 House vote. The operation had two co-leaders. William Daley, a Chicago lawyer and son of the late Mayor Richard J. Daley, had been recruited from his law practice in August 1993 to serve as the White House’s NAFTA point person. Daley had no formal administration title; he operated out of a converted office in the Old Executive Office Building with a small staff and a direct line to the Oval Office. The second co-leader was Bill Richardson, a New Mexico Democrat then serving his sixth term in the House, who had been recruited to whip Hispanic Democrats and Western Democrats whose districts had economic interests in border-region commerce. Daley handled overall strategy and Republican outreach; Richardson handled vote counting and the daily mechanics of district-by-district persuasion.

The whip count as of mid-October 1993 was grim for the administration. Internal Daley counts showed perhaps 95 firm Democratic yeas, perhaps 90 firm Democratic nays, and about 70 Democrats undecided or leaning against. On the Republican side, the count was more favorable but uncertain: perhaps 110 firm Republican yeas, perhaps 30 firm Republican nays, and about 35 Republicans undecided. Reaching the magic number of 218 required pulling perhaps 25 of the 35 undecided Republicans and perhaps 20 to 30 of the 70 undecided Democrats. The math was tight but achievable if the administration could match the right combination of policy concessions to district-specific concerns.

Daley’s strategy through October was to identify each undecided member’s specific economic concern and offer a concession targeted at that concern. The concession menu, assembled through coordination with the Department of Agriculture, the Department of Commerce, and the Office of Management and Budget, included sugar subsidy adjustments for Florida Democrats representing cane-growing districts; orange-juice tariff protections for Florida Republicans representing citrus districts; wheat-export concessions for North Dakota and Kansas members; flat-glass safeguards for West Virginia members representing glass-manufacturing districts; and ceramic-tile protections for Tennessee members representing manufacturing districts. Each concession was negotiated district by district and announced in coordinated press releases timed to maximize political cover for the supporting member. Whether the cumulative cost of these concessions outweighed the trade liberalization benefits of NAFTA itself was a question administration economists privately worried about and publicly never addressed.

The most aggressive Daley negotiations concerned the citrus and sugar provisions for Florida. Florida had twenty-three House members in 1993, fifteen Republicans and eight Democrats, and the state’s agricultural economy was sensitive to Mexican imports of orange juice and sugar. Daley negotiated a fifteen-year tariff phase-out for orange juice rather than the original ten-year phase-out, plus a snap-back provision that would restore tariffs if Mexican imports exceeded specific thresholds. He negotiated sugar quota protections that maintained the existing United States sugar program against Mexican imports for the foreseeable future. The Florida concessions cost the administration substantial leverage with consumer groups, who correctly identified them as protectionist exceptions inside a free trade deal. But the concessions secured roughly seventeen Florida House yes votes that would have been impossible to obtain without them.

Newt Gingrich and the Republican Whip Calculation

The GOP side of the The deal whip operation deserves separate treatment because the Republican yea share of 75.4 percent was not automatic. Newt Gingrich, the House Republican Whip and obvious heir apparent to retiring Minority Leader Bob Michel, faced a calculation that pulled in opposite directions. Free-trade ideology and the Reagan-Bush legacy pulled toward supporting the accord. Pure partisan logic pulled toward letting The President fail on his signature legislative priority. The third pull, harder to articulate but politically important, was Gingrich’s strategic vision for the 1994 midterms and for the longer GOP congressional realignment he had been building since the late 1980s.

Gingrich resolved the calculation by separating The pact from the rest of the Clinton agenda. On The pact specifically, Gingrich would whip aggressively in favor of the deal, deliver the largest possible Republican yea share, and claim partial credit for the policy outcome as a continuation of Reagan-Bush tariff liberalization. On other Clinton priorities, particularly the upcoming health care fight, Gingrich would whip absolute opposition, treating any Republican defections as betrayals of party discipline. The separation served multiple Gingrich goals. It established Republicans as the responsible governing party willing to support sound economic policy when offered, which would help in suburban professional districts where 1992 had been bad for Republicans. It demonstrated Republican legislative effectiveness despite minority status, which would help recruit candidates for 1994. And it positioned Gingrich personally as the architect of the most consequential bipartisan vote of the The President presidency, which would strengthen his standing inside the Republican caucus as Michel’s successor.

The Gingrich whip operation ran parallel to but distinct from the Daley whip operation. Gingrich’s team, working from the Whip’s office in the Capitol’s basement, identified the small number of Republican members who might break ranks on protectionist grounds and applied pressure through committee assignments, fundraising support, and direct personal lobbying. Gingrich himself made roughly forty individual phone calls to wavering Republicans across October and early November 1993. The pressure points he used varied by member: ideological appeals to Reaganite trade-liberalization principles for senior members, district-specific arguments about export opportunities for newer members from agricultural districts, and direct invocations of party discipline for members whose committee seats or leadership positions Gingrich could influence.

The GOP opposition to NAFTA, such as it was, came from a relatively small group of populist or protectionist members concentrated in declining manufacturing districts. Duncan Hunter of California voted no on national-security grounds related to defense industrial base concerns. Helen Delich Bentley of Maryland voted no on protectionist grounds rooted in her shipping-industry background. Pat Buchanan, not yet in Congress but already organizing against NAFTA from his syndicated column, provided rhetorical support for the small Republican opposition without directly mobilizing votes. The forty-three Republican no votes represented roughly a quarter of the Republican caucus, which Gingrich treated as an acceptable defection rate given the ideological diversity of the caucus and the genuine local pressures on members from protectionist districts.

The Daley-Gingrich coordination, while never formalized, was operationally close through October and November. Daley’s team knew which Republicans were undecided; Gingrich’s team knew which Republicans were available for Daley’s policy concessions; the two operations exchanged information through intermediaries multiple times per week. The unusual party-cross coordination was politically defensible only because both sides had reasons to want the agreement passed, but the operational mechanics of cross-party whip cooperation on a The President priority were unprecedented in modern legislative history. The 75.4 percent GOP yea share that resulted was substantially Gingrich’s accomplishment and reflected the strategic logic that Gingrich had imposed on his caucus rather than the natural distribution of Republican views on trade.

Gephardt, Bonior, and the Resistance From Inside the Caucus

The Democratic House leadership’s opposition to The pact was unprecedented in modern presidential politics. Speaker Tom Foley of Washington publicly supported The deal, but with the caveat that he would not whip the caucus on behalf of the deal and would allow members to vote their districts. Majority Leader Richard Gephardt of Missouri actively opposed The pact and announced his opposition in a September 21, 1993 speech to the Center for National Policy that laid out a detailed labor-and-wage-based case against the pact. Majority Whip David Bonior of Michigan went further: he organized a parallel anti-NAFTA whip operation that worked alongside the AFL-CIO and the Citizens Trade Campaign coalition to identify and pressure undecided members.

Gephardt’s opposition was the more important of the two for media purposes. As Majority Leader, Gephardt was the second-ranking Democrat in the Lower chamber and a known presidential aspirant who had run for the nomination in 1988 and would run again in 2004. His September 21 speech and his subsequent op-ed in the New York Times on October 12, 1993, gave political cover to Democratic members who wanted to oppose the deal without appearing to defy the White House. Gephardt’s case rested on three claims: that the side agreements were too weak to address Mexican wage differentials; that the deal would accelerate manufacturing job losses in the industrial Midwest; and that the agreement contained no enforceable currency provisions that would prevent Mexican peso depreciation from undermining the trade balance. Each claim had analytical merit, and Gephardt’s articulation of them gave the anti-NAFTA coalition a more sophisticated public face than Perot’s “giant sucking sound” had provided.

Bonior’s role was more operational. As Majority Whip, Bonior controlled the official Democratic whip operation but could not use that machinery against a sitting Democratic president without unprecedented intra-party rupture. Instead, Bonior built a parallel whip operation, working from his office and from coordination meetings with the AFL-CIO’s legislative office. The Bonior operation tracked roughly the same universe of undecided Democrats that Daley’s operation was tracking and worked to flip each member in the opposite direction. The Bonior operation’s tools were narrower than Daley’s because Bonior did not control White House policy concessions. What Bonior could offer was union political support in the 1994 midterm cycle, AFL-CIO field operations, and the implicit threat of primary challenges in safe Democratic districts where labor was strong. The trade-off the administration faced on each undecided Democrat was a Daley policy concession on one side and a Bonior union threat on the other. In districts where labor was weak or where the White House’s policy concession was specific enough to matter (a sugar provision in a Florida district, a wheat provision in a North Dakota district), the Daley side won. In districts where labor was strong and the Daley concession was generic (Michigan, Ohio, Pennsylvania factory belt), the Bonior side won.

The Findable Artifact: The Final House Vote Breakdown

The November 17, 1993 vote on the implementing legislation, taken at approximately 10:45 p.m. Eastern time after a thirteen-hour debate, recorded the precise distribution that the Daley and Bonior operations had produced. The aggregate count was 234 yeas, 200 nays, and one not voting. The party breakdown is the artifact that defines this article and that any future writer on cross-party presidential coalition-building must reckon with.

Party	Yea	Nay	Pct Yea
Republican (175 total)	132	43	75.4 percent
Democrat (258 total)	102	156	39.5 percent
Independent (1 total)	0	1	0.0 percent
Total (434 voting)	234	200	53.9 percent

The Republican yea share of 75.4 percent and the Democratic nay share of 60.5 percent invert the normal expectation for a Democratic president’s signature initiative. Republicans, the opposition party in the chamber, delivered 132 of the 234 winning yeas. Democrats, the majority party with the President at the head of the ticket, supplied only 102 yeas against 156 nays. Without Republican support, the bill would have failed by a margin of roughly 102 to 332. The administration’s coalition was structurally a Republican coalition with Democratic supplements rather than a Democratic coalition with Republican margin votes. The geographic breakdown reinforces the pattern. Northeastern and Midwestern Democrats voted heavily against NAFTA, reflecting the manufacturing and unions concentrations in those regions. Southern Democrats split more evenly, with Texas, Florida, and North Carolina delivering most of the Democratic yeas from below the Mason-Dixon line. Western Democrats voted slightly more favorably than the caucus average, with California producing a surprisingly mixed result given the state’s manufacturing base. Republican yeas came from across the country, with the heaviest concentration in the Sun Belt and the Plains and the heaviest opposition (such as it was) concentrated in declining-manufacturing districts that had voted Republican on social-conservative grounds despite the economic profile.

The Senate vote three days later, on November 20, 1993, ran along similar but less stark lines. The aggregate count there was 61 yeas to 38 nays, with one not voting. The party breakdown showed 34 Republicans voting yea and 10 voting nay, against 27 Democrats voting yea and 28 voting nay. The Senate’s Democratic margin was narrower, partly because senators represent statewide constituencies where the AFL-CIO concentration in any single congressional district matters less, partly because Senate Democrats from southern and western states with significant cross-border trade had more reason to support the agreement. But even in the Senate, the underlying pattern held: Republicans provided the larger share of the winning coalition.

The Gore-Perot Debate: November 9, 1993

The most-watched single event in the The accord fight was the Larry King Live debate on November 9, 1993, between Vice President Al Gore and Ross Perot. Perot had built much of his 1992 third-party presidential campaign on opposition to NAFTA, and his “giant sucking sound” phrase, delivered in the second presidential debate on October 15, 1992, had become the most-quoted line of his political career. The phrase referred to the sound Perot expected would emerge from American manufacturing employment as production shifted to lower-wage Mexico after The accord’s passage. Through 1993, Perot’s United We Stand America organization had spent roughly fifteen million dollars on anti-The pact advertising and grassroots organizing, and Perot personally had become the public face of opposition in a way that no congressional Democrat including Gephardt could match.

Perot’s organizing infrastructure was unusually sophisticated for a third-party movement. United We Stand America claimed roughly 1.5 million members across all fifty states by mid-1993, with state-level chapters and a Dallas headquarters that functioned as a hybrid between a political action committee, a grassroots organization, and a personal political vehicle for Perot’s continuing public role. The organization’s anti-The accord campaign through 1993 included approximately five thousand local town hall meetings, fifty-five thousand phone calls per week to congressional offices during peak weeks of the legislative fight, a sustained direct-mail program targeting roughly four million households, and a video distribution program that produced roughly five hundred thousand copies of Perot’s anti-NAFTA video for distribution at no cost through local chapters. The organizational scale exceeded that of any single workers union working against The deal and rivaled the AFL-CIO’s aggregate effort. Perot’s personal media platform, which included regular appearances on Larry King Live and his self-funded thirty-minute infomercial program, gave him reach into millions of households that no traditional political organization could match.

The decision to send Gore to debate Perot was not obvious in advance. The administration was acutely aware that the previous October’s Gore-Quayle vice presidential debate had been a wash and that Gore’s reputation as a debater was solid rather than spectacular. Sending the Vice President to confront Perot directly carried real political risk: a Gore loss would consolidate Perot’s anti-The pact position and probably swing additional House Democrats against the pact in the final week before the vote. The decision was made over multiple meetings during the last week of October 1993, with Daley, Stephanopoulos, Mack McLarty, and Gore himself participating. Gore argued for the debate on the grounds that Perot’s debate performance in his 1992 town hall format had been strong but that Perot’s policy expertise was shallow and would not survive sustained substantive engagement. Stephanopoulos and McLarty were skeptical. Clinton’s instinct, as recorded in Bob Woodward’s later reporting, was that Gore could win the debate on substance and that the administration needed a visible event to break Perot’s media monopoly on the opposition case.

Gore’s preparation for the debate was unusually intensive. The Vice President’s office cleared his schedule for the four days before November 9 and conducted three full mock debates with administration staff playing Perot. Robert Reich’s economics team at the Labor Department supplied wage and trade data; the National Economic Council under Robert Rubin supplied macroeconomic projections; the USTR under Mickey Kantor supplied agreement-specific details. Gore’s binders included verbatim quotations from Perot’s previous appearances on Larry King Live, Perot’s 1992 campaign book Save Your Job, Save Our Country, and Perot’s various town hall presentations across 1993. The mock debates identified specific moments where Perot’s previous claims had been factually inadequate or internally contradictory, and Gore drilled repeatedly on the rapid-response framings that would expose those weaknesses without appearing to humiliate Perot personally.

The debate ran for ninety minutes on November 9, 1993, in the CNN Washington studio with Larry King moderating. Gore came prepared with binders of Perot quotations, charts of commerce flow data, and specific challenges on Perot’s earlier claims. Within the first thirty minutes, Gore had pushed Perot into defending several specific factual claims (about Mexican wage levels, about the projected jobs impact of The pact, about the side agreements’ enforcement mechanisms) on which Perot’s responses were incomplete or wrong. Perot’s signature rhetorical style, which depended on confident assertion of populist common sense without sustained substantive backing, did not survive an opponent who had memorized the documentary record. By the end of the debate, Perot had repeatedly lost his temper, had several times refused to answer Gore’s direct questions, and had retreated to his standard talking points in ways that came across on television as evasive rather than principled. Larry King’s even-handed moderation made Perot’s struggle more visible rather than less; King refused to rescue Perot when Gore pressed him on specifics.

The effect of the debate, measured by next-day polling and by Daley’s whip count, was decisive. A CNN-Time poll the morning after gave Gore the debate by a 59 to 32 margin among viewers. A USA Today poll registered a 51 to 26 Gore victory. House Democrats who had been undecided going into the debate broke decisively toward yes in the subsequent week. Daley’s whip count gained roughly twelve additional Democratic commitments between November 10 and November 16, almost all from members who had been wavering and who cited the Perot debate as the moment they decided the agreement was defensible. Bonior’s parallel operation lost no commitments but gained none in the same period. The November 9 debate was the inflection point that turned a narrow win into a comfortable one. Without it, the House margin would likely have been single digits rather than 34 votes.

The post-debate decline of Perot’s anti-NAFTA influence is itself part of the story. United We Stand America’s membership growth, which had been roughly twenty thousand new members per month through the first ten months of 1993, slowed dramatically after November 9. Perot’s media bookings declined as well, with Larry King Live moving from weekly Perot appearances to monthly ones by spring 1994. Perot remained politically active through 1996, when he ran a substantially weaker third-party campaign as the Reform Party nominee, but his stature as the public face of opposition to American commerce liberalization never recovered. The debate transformed Perot from a credible national policy voice into a recurring sideshow whose authority on commerce questions had been visibly undermined. Whether the transformation was deserved or simply a function of Gore’s superior preparation is a question that depends on how one weighs Perot’s broader populist warnings against his specific factual claims, but the political consequence was unambiguous.

The Complication: Was The deal Worth It?

The harder question, and the one this article will not duck, is whether the deal Clinton spent his political capital to pass was worth passing. The economic literature on NAFTA’s effects is unusually divided, partly because the underlying counterfactual question (what would have happened to American manufacturing in the absence of NAFTA) is genuinely difficult and partly because the political stakes of the answer have remained high through three subsequent decades.

Aggregate effects on United States gross domestic product are small in every credible estimate. The Peterson Institute for International Economics calculated in 2014 that The accord had added roughly 0.1 to 0.4 percent to American GDP across the two decades since implementation. The Congressional Research Service’s 2017 review reached a similar range. The aggregate gains from trade liberalization, while positive, were too small relative to the size of the American economy to register as a major macroeconomic event. The agreement was substantively important for specific sectors, particularly automotive, where supply chains became deeply integrated across the United States, Canada, and Mexico, and particularly agriculture, where American grain and meat exports to Mexico grew substantially. But the aggregate effect on the American economy as a whole was modest.

Manufacturing employment is the harder question. The Bureau of Labor Statistics reported manufacturing employment of roughly seventeen million workers in January 1994 when The arrangement took effect; manufacturing employment had declined to roughly twelve million workers by 2017 when USMCA renegotiation began. The five million job decline cannot be cleanly attributed to NAFTA because the same period saw the rise of China as a manufacturing power, the 2001 Chinese accession to the World Trade Organization, the dot-com bust, the 2008 financial crisis, and significant automation in manufacturing processes. Economists who have tried to disaggregate the specific The accord effect from these confounders have reached estimates ranging from roughly 700,000 jobs displaced (the Economic Policy Institute’s 2014 estimate) to roughly 200,000 jobs displaced (the more conservative Peterson Institute estimate). The honest position is that NAFTA contributed measurably to manufacturing employment decline in specific regions, particularly the industrial Midwest and the Southeastern textile belt, while the bulk of manufacturing job losses across the longer period came from forces other than The pact.

Mexican wage outcomes are the most disappointing result. Pre-NAFTA projections by free trade advocates anticipated that Mexican wages would rise substantially as Mexican manufacturing productivity converged on American productivity through investment-flow and supply-chain integration. The actual outcome was different. Mexican wages stagnated through the late 1990s and early 2000s in real terms, partly because the 1994 peso crisis devalued Mexican wages in dollar terms immediately after The accord’s implementation, partly because productivity gains in Mexican manufacturing accrued to capital rather than labor, partly because Mexican labor law enforcement remained weak enough that the side agreements’ unions commission did not deliver substantial improvements in working conditions. The Mexican manufacturing economy grew significantly under NAFTA; Mexican manufacturing employment grew substantially; Mexican wages did not grow proportionally. The compact that had been justified to American labor as one that would eventually equalize wages across the continent did not, in practice, deliver that equalization.

Adam Mayer’s analysis in The Ambivalent Superpower frames the NAFTA outcome as part of a broader 1990s American trade policy that systematically underestimated the costs of liberalization while overestimating the distributive benefits. Mayer argues that the The President administration’s economic team, trained in the neoliberal consensus of the late 1980s and early 1990s, treated aggregate efficiency gains as sufficient justification for trade liberalization and gave inadequate weight to the regional concentration of tariff-related job losses. The backlash that produced Donald Trump’s 2016 The pact-attack campaign, on Mayer’s reading, was a delayed response to the distributive consequences that the Clinton-era trade economists had dismissed as transitional adjustment costs.

Sean Wilentz’s account in The Age of Reagan reaches a partially overlapping but distinct conclusion. Wilentz treats The pact as the signature legislative expression of what he calls “the Clintonian compromise,” in which a Democratic administration accepted the broad outlines of the Reagan-era economic consensus while attempting to soften its harshest distributional consequences through targeted programs. The pact, on Wilentz’s reading, was the most consequential single instance in which The President chose continuity with Reagan-Bush economic policy over the labor-and-progressive demands of his own party’s base. Wilentz’s verdict is harsher than Mayer’s: he treats The accord as the moment when the Democratic Party began the long migration away from its New Deal labor coalition that would culminate in the 2016 election cycle’s working-class realignment.

Joe Klein’s The Natural takes the contrary position. Klein argues that The pact was the right policy for the wrong constituency: that the agreement’s modest aggregate gains were real, that the side pacts were the best unions and environmental protections politically achievable in 1993, and that The White Chamber’s willingness to pursue the agreement against his own party’s opposition was the kind of cross-coalition governance that the country needed more of, not less. Klein concedes the distributive consequences in specific regions but argues that those consequences were preventable through adequate adjustment assistance and worker retraining, which the United States, unlike Western European countries facing similar trade disruptions, did not provide at scale. Klein’s NAFTA verdict, in short, is that the agreement was correct and the United States failed to make the complementary domestic policy investments that would have made the deal work for American workers.

Jeffrey Cohen’s Presidential Leadership in Public Opinion offers a fourth analytical frame that focuses on the agreement’s effect on The White Chamber’s longer political coalition rather than on the agreement’s economic merits. Cohen argues that NAFTA was the first major instance in which Clinton’s presidential leadership style, which depended on cross-coalition settlement-making and on symbolic centrist positioning, came into open conflict with his party’s interest-group coalition. Cohen’s analysis traces how the November 1993 vote reshaped subsequent The President-Democratic Party relations, particularly with organized labor, and shows that the cost of the The pact fight registered on Clinton’s coalition before it registered on his approval ratings or on the broader American economy. The 1994 midterm losses, which delivered GOP control of both chambers of Congress for the first time since 1954, can be partially traced to depressed Democratic turnout in industrial-belt districts where unions had been most active in opposing The deal. Cohen’s account complicates the Klein verdict by showing that the cost of cross-coalition governance was higher than Klein acknowledges, while complicating Wilentz’s verdict by showing that the cost was specific and concentrated rather than broadly transformative of the Democratic coalition’s structure.

Peter Baker’s reporting in The Breach: Inside the Impeachment and Trial of William Jefferson Clinton provides the immediate political context for understanding the longer Clinton-Democratic relationship as it played out across the second term. Baker frames the impeachment-era Democratic loyalty to Clinton as the eventual repayment of political debts that the The arrangement-era Democratic Party had felt Clinton had defaulted on. The Democratic House caucus’s near-unanimous opposition to impeachment in December 1998, on Baker’s reading, reflected not just opposition to Republican overreach but also a settled understanding that the party would protect its president after he had survived multiple internal disagreements including the 1993 The deal fight, the 1996 welfare reform vote, and the 1998 IMF funding vote. The cross-coalition governance template that produced the 234 to 200 The deal outcome eventually exhausted its political credit by the late 1990s, and the impeachment period saw the Democratic caucus reconsolidate around the President in a way it had not done on legislative priorities throughout the previous five years.

Adjudicating among Klein, Mayer, Wilentz, Cohen, and Baker is not the task of this article, because the question they disagree about (the long-run wisdom of trade liberalization) is genuinely contested and the evidence underdetermines a verdict. What the historical record does support clearly is the more limited claim that Clinton in 1993 made a coalition-building choice that no other president of the late twentieth century made on a comparable scale, that the choice succeeded on its immediate procedural terms, and that the choice’s distributional consequences played out across three subsequent decades in ways that the administration did not fully anticipate. The decision was made, the vote was won, and the political consequences arrived on schedules longer than the electoral cycle that ratified or rejected the decision.

The Labor Coalition That Failed to Defeat NAFTA

The AFL-CIO’s anti-NAFTA campaign through 1993 was the largest single legislative fight the AFL-CIO federation had undertaken since the 1978 workers law reform effort that the Carter administration had failed to pass. Lane Kirkland, the AFL-CIO president since 1979, treated NAFTA as an existential question for the federation’s manufacturing-sector affiliates. The campaign included direct lobbying, member mobilization, advertising, and coalition-building with environmental groups, consumer advocates, and the Citizens Trade Campaign coalition led by Public Citizen’s Lori Wallach. The aggregate organizational effort exceeded fifty million dollars across all participating organizations and reached an estimated thirty million American households through television advertising alone. By any measure of resources deployed, the labor coalition was running the largest anti-The pact operation in American legislative history.

The coalition’s failure to defeat The pact despite this scale is a structural lesson in the limits of single-interest legislative campaigns. The first structural problem was the geographic concentration of labor’s political strength. Union density in 1993 was concentrated in northeastern and Midwestern states whose congressional delegations were already heavily Democratic and already leaning anti-NAFTA. The marginal value of additional union pressure in those districts was low because the members were already committed. In contrast, the swing districts that would determine the vote were concentrated in southern, western, and suburban areas where union density was below ten percent and where unions’s mobilization capacity was correspondingly weak. The AFL-CIO could not deliver pressure where pressure was most needed.

The second structural problem was the environmental coalition split. The Kantor side pacts, criticized by AFL-CIO as inadequate, were treated as adequate by the Natural Resources Defense Council, the National Audubon Society, and the Environmental Defense Fund. The split inside the environmental movement meant that the anti-The accord coalition could not credibly claim to represent the entire progressive policy community. Members from suburban districts where environmental groups were politically significant could point to the NRDC and Audubon endorsements as cover for supporting the agreement. The administration had identified this fault line months in advance and had specifically courted the moderate environmental groups with side-deal language that addressed their priority concerns about cross-border pollution and species protection.

The third structural problem was the absence of a positive labor agenda that could compete with the tariff-liberalization frame. Labor’s case against NAFTA was substantially negative: the deal would destroy jobs, the side agreements were inadequate, the long-term wage convergence would not happen. What labor lacked was a positive trade-policy alternative that could be articulated to swing members as the responsible policy choice. The closest thing to a labor alternative was the fair-trade framework promoted by Citizens Trade Campaign, which would have conditioned commerce liberalization on demonstrable labor and environmental standards in trading partner countries. The fair-trade framework was substantively serious but operationally impossible in 1993 because it would have required reopening the NAFTA text with Mulroney and Salinas, neither of whom would have agreed to the kind of binding unions and environmental conditions that fair-tariff advocates wanted. The AFL-CIO coalition therefore found itself defending a status quo (pre-NAFTA tariff structure) rather than offering an alternative forward-looking framework, which left swing members without a positive narrative for opposition.

The post-vote consequences for labor’s political position were significant. The AFL-CIO’s relationship with the The White House administration remained strained for the remainder of the first term. The federation’s political endorsements in 1994 included threats of primary challenges against pro-The deal House Democrats in twelve specific districts, although most of those threats did not materialize into actual primary contests. The 1995 election of John Sweeney to replace Kirkland as AFL-CIO president was driven partly by internal federation frustration with Kirkland’s perceived inadequate response to The accord’s passage, and Sweeney’s New Voice slate explicitly promised more aggressive political mobilization on trade and other priorities. The longer trajectory of organized unions’s declining political influence through subsequent decades has multiple causes, but the visible defeat in November 1993 marked the beginning of a period in which worker could not reliably deliver votes on its highest-priority legislative fights despite substantial organizational investment.

The Night of November 17: A Moment-in-Time Reconstruction

The Lower chamber debate on the implementing legislation began at 9:48 a.m. on November 17, 1993, under a special rule allowing thirteen hours of debate equally divided between proponents and opponents. The format reflected the magnitude of the vote: most major legislative debates in 1993 ran four to six hours, and the thirteen-hour allocation signaled both sides’ recognition that the chamber’s deliberation deserved an unusual record. The floor managers were Sam Gibbons of Florida and Bill Archer of Texas for the proponents, working from the Ways and Means Committee where Gibbons chaired and Archer was ranking Republican. The opposition floor managers were David Bonior of Michigan and Marcy Kaptur of Ohio.

The morning hours produced predictable speeches from both sides. Gephardt’s prepared remarks ran twenty-eight minutes and laid out the wage-differential and manufacturing-employment case against the pact. Newt Gingrich’s remarks ran twenty-three minutes and made the free-trade economic case in characteristically aggressive terms. Charles Rangel of New York, who supported the agreement despite representing a district with significant industrial employment, gave a fifteen-minute speech that other administration supporters cited approvingly through the afternoon. Pat Schroeder of Colorado, who opposed the deal primarily on labor grounds, gave a speech that focused on women workers in industrial whose displacement she predicted would be disproportionate.

The afternoon session ran from 1:30 p.m. through 7:00 p.m. with a brief recess for committee meetings unrelated to NAFTA. By 5:00 p.m., the C-SPAN broadcast had drawn unusually large viewership for a midweek legislative session, with estimated audience peaks above one million households. Daley’s whip count by 6:00 p.m. showed 218 to 220 firm yeas, narrowly above the 218-vote majority threshold, with another fifteen members listed as leaning yes but not yet firmly committed. Bonior’s count showed 195 to 200 firm nays with another ten members listed as leaning no. The remaining undecided members, perhaps twenty in total, were being worked aggressively by both sides through the dinner hour.

The evening session ran from 7:30 p.m. through the 10:45 p.m. vote. The most-watched moment of the evening came at approximately 8:30 p.m. when Bill Richardson, the Daley operation’s vote counter, came to the floor with a manila folder containing the final concession packages for four specific members whose votes the White House had been negotiating through the previous week. Two Florida Democrats from sugar-producing districts, one West Virginia Democrat from a glass-manufacturing district, and one Indiana Democrat from a steel-producing district received their concession announcements during the evening session. All four voted yes when the roll was called. The Bonior operation, watching Richardson’s movements on the floor, recognized that the late concessions had likely pushed the vote count beyond recovery. Bonior himself voted no without making a closing speech; the Whip’s office released a brief statement registering his opposition and pledging to continue the fight on subsequent trade legislation.

The vote itself began at 10:36 p.m. with the Speaker’s gavel and ran fifteen minutes under the standard House voting procedure. The electronic display board in the chamber showed the running count in real time, and the room grew progressively quieter as the yes count climbed past 200 and then past 218. The final tally of 234 to 200 was announced at 10:51 p.m. by acting chair Charles Rose of North Carolina. The President, watching the vote on television in the White House residence with Hillary Clinton, Al Gore, and senior staff, made congratulatory phone calls to Daley, Richardson, Gingrich, and Speaker Foley within the next ninety minutes. The administration’s celebration was muted by the recognition that the Senate vote three days later would test whether the cross-coalition strategy could be sustained or whether House Democrats would punish the administration on subsequent priorities. The Senate vote of 61 to 38 on November 20 confirmed the strategy’s success but did not end the longer political fight over the pact’s distributional consequences.

The morning newspapers of November 18 led with the vote count and the unusual party breakdown. The New York Times front-page story by David Rosenbaum emphasized the Republican margin of victory: 132 Republican yeas had carried a Democratic president’s signature legislation. The Washington Post’s coverage by Helen Dewar described the vote as “a victory built on Republican votes” and quoted Gingrich’s post-vote statement that NAFTA was “a triumph of policy over partisanship.” The Wall Street Journal’s editorial board praised both Clinton and Gingrich while noting that the longer electoral fallout would depend on whether the agreement’s economic benefits arrived faster than its concentrated regional costs. Within forty-eight hours of the vote, both sides had moved on to the next legislative fight: health care reform, where the cross-coalition strategy that had worked for The accord would not be available because no Republican constituency had any interest in delivering The President a second signature legislative victory.

The Verdict

The InsightCrunch claim on the NAFTA passage, the namable thesis this article advances, is this: Clinton’s NAFTA was the most consequential single instance of cross-coalition presidential governance in the second half of the twentieth century, and the success of that strategy depended on three specific operational decisions whose combination has not been replicated since.

The first decision was the August 1993 side-agreement strategy. By producing parallel worker and environmental agreements rather than reopening the underlying NAFTA text, Kantor preserved the December 1992 deal that Bush had signed while giving Clinton a substantive policy claim that he had improved the settlement materially. The strategy lost the AFL-CIO and gained the moderate environmental groups, splitting the opposition coalition along a fault line that the administration had identified and exploited. Without the side agreements, the pact would have remained a Bush accord that Clinton was merely ratifying, with no policy distance to claim and no environmental cover to gather.

The second decision was the September 14 four-president ceremony. By placing Bush, Ford, and Carter on stage with him, The President converted NAFTA from a partisan controversy into a centrist consensus initiative. The signaling effect, particularly on swing House Democrats and undecided House Republicans, was substantial. The ceremony also gave Bush a graceful path to support the agreement publicly, which made it harder for Lower chamber Republicans to oppose Clinton on the grounds that they were defending Bush administration legacy.

The third decision was the November 9 Gore-Perot debate. By engaging Perot directly on substantive ground, Gore demonstrated that the opposition case rested on confidently asserted but factually inadequate claims. The debate’s polling effect translated directly into the dozen undecided Democratic commitments that produced the comfortable rather than narrow House margin. Without the debate, the November 17 vote would have been substantially closer and would have left less room for subsequent administration priorities.

The three decisions together produced the 234 to 200 outcome. The pattern they collectively established, of cross-aisle coalition-building anchored in centrist symbolism and substantive debate engagement, is the model the article names as the InsightCrunch coalition governance template. The template has obvious historical predecessors and obvious historical successors, but its 1993 application is the cleanest single case for analysis. John Adams’s 1800 refusal to escalate the Quasi-War with France, which broke with his own Federalist Party’s hawks and contributed materially to his electoral defeat that November, is the eighteenth-century forerunner. Gerald Ford’s September 1974 pardon of Richard Nixon, which broke with the Republican Party’s preference for prolonged prosecution and cost Ford the 1976 election, is the postwar parallel. George H. W. Bush’s 1990 budget agreement with congressional Democrats, which broke with his “no new taxes” campaign pledge and contributed materially to his 1992 defeat, is the immediate predecessor. The pattern across these cases is consistent: presidents who govern across coalition often pay price, and the price arrives on cycles longer than the immediate election that ratifies the decision. The President survived the immediate cycle because The pact’s distributional consequences took years to materialize, but the Democratic Party’s long-term realignment along trade-policy lines, culminating in 2016, can be read as the delayed bill for the November 17, 1993 vote.

Legacy: From The pact to USMCA and Beyond

The 2017 to 2020 renegotiation of The deal into the United States-Mexico-Canada Agreement is the closing argument on the 1993 passage. Donald Trump campaigned in 2016 on explicit attacks against NAFTA as “the worst trade deal ever made” and entered office in January 2017 committed to renegotiation or withdrawal. The renegotiation process, conducted by United States Trade Representative Robert Lighthizer between 2017 and 2018, produced an updated agreement that retained roughly 90 percent of the original NAFTA structure while adding stronger labor enforcement provisions, country-of-origin rules for automotive manufacturing that favored North American supply chains over Asian inputs, and a sunset clause requiring periodic renegotiation. USMCA passed Congress in January 2020 with substantial bipartisan support, including the votes of many Democrats who had opposed NAFTA in 1993. The renegotiated deal was, in substantive terms, a modest revision rather than a fundamental redesign. The political reframing, however, was significant: USMCA was sold to American voters as a NAFTA replacement rather than a NAFTA continuation, and the Trump administration claimed credit for renegotiating away the harms that The White House had imposed.

The deeper legacy of the 1993 passage is the realignment it accelerated. The Democratic Party in 1993 was still dominated by an industrial-state, labor-aligned coalition whose policy preferences ran toward protectionism and managed trade. The party in 2024 is substantially different: more dependent on professional-class, educated suburban voters whose preferences run toward globalization and away from industrial protection. The Republican Party in 1993 was the consistent free-trade party whose Reagan-Bush trade liberalization legacy NAFTA continued. The Republican Party in 2024 is substantially different: more dependent on working-class voters in regions hit hardest by trade-related manufacturing declines, more sympathetic to industrial policy and tariff protection. The two parties have substantially traded places on trade since 1993, and the realignment can be tracked through the slow accumulation of working-class Democratic defections in Midwestern and Pennsylvanian counties that voted for Clinton in 1992 and 1996, drifted toward Republicans across the 2000s, and broke decisively Republican in 2016. The deal did not cause the realignment, but it is among the structural events that made the realignment possible.

For the house thesis of this series, the imperial presidency forged in four crises (Civil War, Great Depression, World War II, Cold War) that outlived each emergency, The deal threads only lightly. The accord was not an exercise in executive power expansion; it was an exercise in executive coalition-building across party lines on policy ground that Congress retained substantial authority over. The The President presidency’s contribution to executive power expansion runs through other moments (the Kosovo intervention without congressional authorization, the use of executive orders to set environmental policy, the line-item veto’s brief 1996 experiment) rather than through The pact. What The accord does illustrate, however, is the alternative model: presidents who choose to build coalitions across party lines rather than expand executive prerogative to govern without those coalitions. The two models have coexisted across the postwar period, and the Clinton 1993 case is among the cleanest examples of the coalition path. Whether subsequent presidents have followed the coalition path more or less than the prerogative-expansion path is a question this series will examine across multiple cases. The 1993 NAFTA decision is the benchmark against which other cross-coalition governance attempts can be measured.

The cross-links from this article should be read as instructions for following the pattern into adjacent cases. The President’s 1994 Rwanda non-intervention, examined in the companion article on the Rwanda decision, shows the cost side of the Clinton coalition-building approach: foreign policy initiatives that lacked domestic coalition often went unpursued. Clinton’s distributed credit with Gingrich on the late-1990s balanced budget, examined in the budget myth-bust article, shows the same coalition-governance template applied to fiscal policy rather than trade. John Adams’s 1800 refusal to escalate the war with France, examined in the Adams decision article, provides the eighteenth-century template that Clinton’s 1993 vote echoes structurally. George H. W. Bush’s 1990 budget deal, examined in the Bush Sr. budget article, provides the immediate parallel case of a president breaking with his own party’s base on economic policy. Each of these cases sits inside the broader pattern that the InsightCrunch coalition governance template names.

Frequently Asked Questions

Q: Who negotiated the original NAFTA agreement?

The original North American Free Trade Agreement was negotiated by the George H. W. Bush administration between June 1991 and August 1992. The United States Trade Representative Carla Hills led the American negotiating team. Jaime Serra Puche led the Mexican team for President Carlos Salinas de Gortari, and Michael Wilson led the Canadian team for Prime Minister Brian Mulroney. The agreement was concluded in August 1992 and signed by the three heads of state on December 17, 1992, between Bush’s November electoral defeat and The White House’s January inauguration. Bush’s decision to sign before leaving office created a fait accompli for the incoming Clinton administration, which inherited the signed pact and had to choose whether to push its implementing legislation through Congress, renegotiate it, or abandon it. Clinton chose to push it through after supplementing it with side pacts on workers and environment that the Bush negotiators had not included.

Q: Why did Clinton support NAFTA despite Democratic opposition?

The President supported NAFTA for a combination of policy and political reasons. On policy, his economic team, led by Robert Rubin at the National Economic Council and Lloyd Bentsen at Treasury, believed that trade liberalization would produce modest aggregate gains for the American economy while strengthening and security ties to Mexico during the country’s transition from one-party authoritarian rule. On political grounds, Clinton believed that demonstrating willingness to pursue economic policy against his own party’s AFL-CIO-aligned base would establish his New Democrat credentials with suburban professional voters whose support he needed for 1996. He also calculated, correctly, that the immediate political cost of The deal would fall mostly on House Democrats from factory districts rather than on the presidency itself, since trade votes traditionally produce delayed rather than immediate electoral consequences.

Q: What were the NAFTA side agreements on labor and environment?

The North American Compact on Labor Cooperation and the North American Agreement on Environmental Cooperation were parallel pacts negotiated by Mickey Kantor’s USTR office between May and August 1993 and signed alongside the original NAFTA text in September 1993. The unions deal established a trinational Commission for Labor Cooperation headquartered in Dallas with authority to investigate alleged non-enforcement of each country’s domestic workers laws and, in limited circumstances, impose monetary penalties and trade sanctions. The environmental agreement established a Commission for Environmental Cooperation in Montreal with similar authority over environmental law enforcement. Both pacts were criticized by labor and environmental groups as too weak to address the underlying wage and regulatory differentials driving concern, but they provided political cover for the moderate Democrats and centrist environmental organizations whose support the administration needed.

Q: What was Ross Perot’s “giant sucking sound” warning?

Ross Perot’s famous phrase came in the second presidential debate on October 15, 1992, when Perot warned that NAFTA would produce “a giant sucking sound going south” as American manufacturing jobs migrated to lower-wage Mexico. The image became the most-quoted line of Perot’s political career and the centerpiece of his United We Stand America organization’s anti-The pact campaign through 1993. Perot spent roughly fifteen million dollars of his own money on anti-The pact advertising and grassroots organizing through the year. The phrase captured a real concern about wage-arbitrage-driven manufacturing migration but oversimplified the underlying economic dynamics. Manufacturing employment did decline significantly in subsequent decades, but most of the decline came from automation, Chinese competition after 2001, and broader globalization rather than from Mexican migration specifically. Perot’s warning was directionally correct on manufacturing employment but substantially wrong on the specific NAFTA-Mexico channel he emphasized.

Q: Who won the Gore-Perot debate on November 9, 1993?

Al Gore decisively won the Larry King Live debate against Ross Perot on November 9, 1993. The debate ran ninety minutes on CNN with Larry King moderating. Gore came prepared with binders of Perot quotations and trade data and pushed Perot into defending specific factual claims that Perot’s responses did not adequately address. By the end of the debate, Perot had repeatedly lost his temper and retreated to talking points in ways that came across as evasive rather than principled. Next-day polling showed Gore winning by margins ranging from 27 to 33 points depending on the survey. The debate’s political effect was substantial: roughly twelve undecided House Democrats broke decisively toward yes in the week between the debate and the November 17 vote, almost all citing the debate as the moment they decided the agreement was defensible. Without the debate, the Chamber margin would likely have been single digits.

Q: What was the House vote on The accord?

The Chamber of Representatives passed the NAFTA implementing legislation on November 17, 1993, by a vote of 234 to 200, with one member not voting. The party breakdown was unusual: 132 Republicans voted yes and 43 voted no, while 102 Democrats voted yes and 156 voted no. The lone Independent, Bernie Sanders of Vermont, voted no. Republican members provided 56 percent of the winning yea coalition despite being the minority party in the chamber. The vote came at approximately 10:45 p.m. Eastern time after a thirteen-hour debate. The geographic distribution showed heavy Democratic opposition from Northeastern and Midwestern manufacturing districts, more evenly split Democratic votes from the South, and slightly more favorable Democratic votes from the West. GOP yeas came from across the country, with the heaviest concentration in the Sun Belt and Plains states.

Q: Who provided the majority of votes for The deal passage?

Republicans provided the majority of the winning coalition for NAFTA. Of the 234 yea votes in the Chamber, 132 came from Republicans and 102 from Democrats, meaning Republicans supplied 56 percent of the winning margin. Without Republican support, the bill would have failed by approximately 102 to 332. This party breakdown inverted the standard expectation for a Democratic president’s signature legislation, where the majority party typically supplies the bulk of yea votes and the minority provides marginal support. The unusual coalition pattern resulted from a combination of Democratic unions-base opposition led by Majority Leader Gephardt and Majority Whip Bonior, Republican free-trade ideological support orchestrated by Whip Newt Gingrich, and a White House whip operation led by William Daley that prioritized Republican outreach after concluding that the Bonior-aligned Democratic bloc was unrecoverable.

Q: Did The pact create or destroy American jobs?

The accord’s net employment effect remains contested among economists three decades after passage. Aggregate models from the Peterson Institute for International Economics and the Congressional Research Service estimate small net positive effects on American GDP (roughly 0.1 to 0.4 percent) and ambiguous net effects on employment. Regional models tell a different story: industrial employment in specific districts of the industrial Midwest and the Southeastern textile belt declined more rapidly after The deal than comparable regions less exposed to Mexican trade competition. Estimates of The deal-specific job displacement range from roughly 200,000 (Peterson Institute) to 700,000 (Economic Policy Institute) across the first fifteen years of the deal. The honest answer is that NAFTA contributed measurably to factory job losses in specific regions while the bulk of American manufacturing employment decline across the longer period came from automation, Chinese competition after 2001, and broader globalization forces.

Q: What did NAFTA do to manufacturing employment?

American manufacturing employment was approximately 17 million workers when NAFTA took effect in January 1994 and had declined to approximately 12 million workers by 2017 when USMCA renegotiation began, a net decline of 5 million jobs across 23 years. Attributing specific shares of this decline to The accord is difficult because the same period saw Chinese accession to the WTO in 2001, two major recessions, and substantial automation in manufacturing processes. Economists who have attempted to disaggregate the NAFTA effect from these confounders estimate that NAFTA contributed roughly 200,000 to 700,000 of the total decline, depending on methodology. The deal’s effects were concentrated in specific industries (apparel, textiles, some automotive subsectors) and specific regions (Midwest, Southeast). Other manufacturing subsectors (advanced manufacturing, aerospace, defense) were largely unaffected or modestly positively affected by deeper supply-chain integration with Mexico and Canada.

Q: How did The accord affect Mexican wages?

Mexican wage outcomes under NAFTA were the most disappointing result for free trade advocates who had predicted that trade liberalization would substantially raise Mexican living standards. Real Mexican wages stagnated through the late 1990s and early 2000s, partly because the 1994 peso crisis devalued Mexican wages in dollar terms immediately after NAFTA’s implementation. Mexican manufacturing employment grew substantially under the agreement, but productivity gains accrued primarily to capital rather than labor, and Mexican labor law enforcement remained weak enough that the side agreement’s labor commission did not deliver substantial improvements in working conditions or wages. The Mexican manufacturing economy grew significantly under NAFTA in absolute terms, but the predicted wage convergence with the United States did not occur, contributing to continued migration pressure across the border and to the case for the stronger labor provisions eventually included in the 2020 USMCA.

Q: What is USMCA and how does it differ from The accord?

The United States-Mexico-Canada Agreement is the renegotiated trade agreement that replaced The pact in July 2020 after negotiations conducted by the Trump administration’s United States Trade Representative Robert Lighthizer between 2017 and 2018. USMCA retains roughly 90 percent of the original The accord structure but adds stronger labor enforcement provisions (with rapid-response mechanisms allowing complaints against specific facilities rather than only against general non-enforcement patterns), updated country-of-origin rules for automotive manufacturing that require higher North American content for tariff-free treatment, new digital commerce provisions absent from the 1993 text, and a sunset clause requiring periodic six-year review and renegotiation. In substantive terms, USMCA is a modest update rather than a fundamental redesign. The political reframing, however, was significant: USMCA was sold to American voters as a NAFTA replacement rather than a continuation, and the Trump administration claimed credit for renegotiating away the harms that Clinton had imposed in 1993.

Q: Who were the key Democratic opponents of The deal?

The key Democratic opponents of NAFTA in the House were Majority Leader Richard Gephardt of Missouri and Majority Whip David Bonior of Michigan. Gephardt’s September 21, 1993 speech to the Center for National Policy laid out the substantive case against the agreement on unions and wage grounds, and his October 12 New York Times op-ed gave political cover to Democratic members who wanted to oppose without appearing to defy the White House. Bonior organized a parallel anti-The accord whip operation that worked alongside the AFL-CIO’s legislative office and the Citizens Trade Campaign coalition to identify and pressure undecided members. Other significant Democratic opponents included Marcy Kaptur of Ohio, John Conyers of Michigan, and Sherrod Brown of Ohio. In the Senate, Democratic opposition was led by Donald Riegle of Michigan and Ernest Hollings of South Carolina. Union union opposition was led by Lane Kirkland of the AFL-CIO and Owen Bieber of the United Auto Workers.

Q: What was Mickey Kantor’s role in passing The deal?

Mickey Kantor was the United States Trade Representative who negotiated the August 1993 side agreements on labor and environmental cooperation that made The accord politically viable for The President. Kantor had managed The President’s 1992 campaign and arrived at USTR with no prior trade specialization, but he proved an effective negotiator with Jaime Serra Puche and Michael Wilson on the side agreements. The agreements established trinational commissions with limited enforcement authority over alleged non-enforcement of each country’s domestic labor and environmental laws. The commissions were criticized as too weak by the AFL-CIO and the more aggressive environmental groups, but they provided sufficient political cover for moderate Democrats and centrist environmental organizations to support the underlying agreement. Kantor’s success in producing the side agreements without reopening the NAFTA text itself was the critical operational achievement that distinguished The President’s deal from a pure ratification of the Bush arrangement.

Q: Why did Republicans support a Democratic president’s trade deal?

House Republicans supported The accord for ideological and strategic reasons. Ideologically, the Republican Party in 1993 was the consistent free-trade party whose Reagan-Bush trade liberalization legacy The pact continued. Most Lower chamber Republicans had supported fast-track authority in 1991 and were on record as favoring tariff liberalization on free-market principles. Strategically, Whip Newt Gingrich calculated that allowing Clinton a foreign policy success would not damage GOP prospects in 1994, particularly because Clinton was already paying significant political costs with labor and the progressive base for pursuing the agreement. Gingrich also believed that demonstrating Republican willingness to support sound economic policy when offered would strengthen the party’s claim to responsible governance. Some House Republicans, particularly from industrial-heavy districts in the upper Midwest, opposed The accord on protectionist grounds, but the 75.4 percent Republican yea share showed that ideological support overwhelmed protectionist resistance within the GOP caucus.

Q: How did The accord shape Clinton’s presidency politically?

NAFTA shaped The President’s first-term politics in three significant ways. First, it consumed substantial political capital that the administration had hoped to spend on health care reform; the November 1993 The pact fight ran in parallel with the Hillary Clinton-led health care task force, and the administration’s inability to muster Democratic legislative discipline for NAFTA foreshadowed the broader Democratic legislative dysfunction that would kill health care reform in 1994. Second, it cemented The President’s reputation with suburban professional voters as a New Democrat willing to pursue policy against labor and progressive opposition, which contributed to his 1996 reelection coalition. Third, it generated lasting resentment within the labor wing of the Democratic Party that contributed to depressed Democratic turnout in the 1994 midterms and to the broader Democratic loss of working-class support across subsequent decades. The compact was a procedural success and a longer-run cost whose distributional consequences arrived on cycles longer than Clinton’s own presidency.

Q: What was the Senate vote on The accord?

The Senate passed the NAFTA implementing legislation on November 20, 1993, by a vote of 61 to 38, with one senator not voting. The party breakdown was 34 Republicans yes and 10 no, against 27 Democrats yes and 28 no. The Senate’s Democratic margin was narrower than the Lower chamber’s, partly because senators represent statewide constituencies where the workers concentration in any single congressional district matters less, partly because Senate Democrats from southern and western states with significant cross-border trade had more reason to support the agreement. Democratic yes votes came primarily from western and southern senators including Bentsen of Texas, Hollings’s eventual replacement reasoning notwithstanding, and senators from non-manufacturing states. Democratic no votes came from northeastern and Midwestern manufacturing-state senators including Riegle of Michigan, Metzenbaum of Ohio, and Hollings of South Carolina. GOP opposition came from a small group of conservative populist senators who opposed the agreement on sovereignty grounds.

Q: Was The accord constitutional?

NAFTA was constitutional under standard treaty-and-tariff-deal jurisprudence. The agreement was passed as a congressional-executive pact rather than as a Senate-ratified treaty, using the fast-track authority Congress had granted Bush in 1991. The fast-track procedure required Congress to vote up-or-down on implementing legislation without amendments, a procedure that the Supreme Court had not specifically addressed for NAFTA but that had been used for prior trade pacts without successful constitutional challenge. Some constitutional scholars, including Bruce Ackerman and David Golove, argued in the mid-1990s that major commercial treaties should require two-thirds Senate ratification under the Treaty Clause rather than simple majority passage under the Commerce Clause, but their position did not prevail. The Made in the USA Foundation challenged The pact’s constitutional status in federal court in 1998, but the Eleventh Circuit dismissed the case on standing grounds in 2001 without reaching the constitutional question.

Q: How did unions unions respond to The pact?

American unions unions opposed NAFTA aggressively and consistently from negotiation through passage. The AFL-CIO under Lane Kirkland committed substantial financial and organizational resources to defeating the agreement, including approximately twenty million dollars in advertising and grassroots organizing through 1993. The United Auto Workers, the Steelworkers, the Teamsters, and the Communications Workers of America were the most active individual unions. Worker’s opposition rested on three claims: that Mexican wage differentials would drive American manufacturing job migration; that the side pacts were too weak to address labor enforcement problems; and that NAFTA’s investor-state dispute provisions would constrain American regulatory authority over foreign-owned manufacturing facilities. After the November 17 defeat, unions leaders publicly committed to opposing the The President administration on subsequent priorities and to supporting Democratic primary challenges in 1994 against pro-The pact Democrats. The AFL-CIO-Democratic Party relationship did not fully recover for the remainder of the Clinton presidency.

Q: What did Carter, Bush, and Ford say at the signing ceremony?

At the September 14, 1993 East Room ceremony, all three former presidents delivered remarks supporting NAFTA in distinct registers. Jimmy Carter spoke briefly and somewhat defensively, focusing on the labor and environmental side agreements as the additions that made the underlying deal worthy of progressive support. Gerald Ford spoke in characteristically blunt terms about the economic case for tariff reduction and the importance of free-trade ideology for postwar American prosperity. George H. W. Bush gave the longest and most generous speech, praising The President’s “courage” in pursuing the pact against the wishes of his own party and recasting the December 1992 signing as a bipartisan accomplishment that Clinton had improved with the side agreements. Bush’s willingness to share the stage with the man who had defeated him ten months earlier was the most striking element of the ceremony. The four-president tableau was preserved in major newspaper photography for the following morning’s coverage and served the administration’s framing of NAFTA as a centrist consensus initiative.

Q: How does The deal compare to other cross-party presidential deals?

NAFTA sits within a small set of cases where presidents built legislative coalitions across party lines against their own party’s leadership. The closest parallel is George H. W. Bush’s 1990 budget agreement, which broke with his “no new taxes” campaign pledge and required Democratic congressional support against Republican opposition led by Newt Gingrich. John Adams’s 1800 refusal to escalate the Quasi-War with France against his own Federalist Party’s hawks is the eighteenth-century precedent. Gerald Ford’s September 1974 pardon of Richard Nixon, which broke with Republican preference for prolonged prosecution, is the postwar parallel. Each of these cases produced significant short-term political costs for the president who chose the cross-coalition path; Adams lost in 1800, Ford lost in 1976, Bush Sr. lost in 1992. Clinton survived the 1996 cycle because The deal’s distributional consequences took years to materialize, but the Democratic Party’s long-term realignment along trade-policy lines, culminating in 2016, can be read as the delayed bill for the November 17, 1993 vote.

Q: Did NAFTA help or hurt American consumers?

NAFTA produced modest aggregate consumer benefits through lower prices on imported goods, particularly fresh produce, apparel, and certain manufactured items, while imposing concentrated job losses on workers in trade-exposed manufacturing sectors. The Peterson Institute for International Economics estimated in 2014 that American households gained roughly 200 to 400 dollars per year on average through lower prices attributable to The accord, with higher gains for lower-income households who spent larger shares of income on tradeable goods. These aggregate consumer gains were diffuse and largely invisible to individual households, while the manufacturing job losses were concentrated, visible, and politically organized. The asymmetry between diffuse consumer benefits and concentrated worker costs is a recurring pattern in trade policy and helps explain why The pact’s opponents have remained more effective than its political defenders despite the aggregate consumer gains being larger than the aggregate worker losses in standard economic models.

Q: What lessons did later administrations draw from the NAFTA fight?

Later administrations drew several lessons from the 1993 NAFTA passage. The The President administration itself shifted to more cautious trade negotiations after 1993, pursuing the Uruguay Round of GATT and the resulting World Trade Organization in 1994 through 1995 but declining to push for additional regional agreements at the scale of The pact. The George W. Bush administration pursued bilateral trade pacts with smaller partners (Chile, Singapore, Australia, the Central American countries through CAFTA) rather than attempting another large regional deal, partly because the political costs of The pact had clarified the diminishing marginal returns of commerce liberalization once the major trading partners were already covered. The Obama administration pursued the Trans-Pacific Partnership through 2010 to 2016 but was unable to secure congressional approval before the 2016 election, and the Trump administration withdrew the United States from TPP in January 2017. The political experience of The deal contributed substantially to the broader decline of American commerce-liberalization ambition across the three subsequent decades.

Q: How did Mexico benefit from NAFTA?

Mexico benefited from The pact in several measurable dimensions while falling short of pre-implementation expectations in others. Mexican manufacturing exports to the United States grew from approximately 40 billion dollars in 1993 to approximately 300 billion dollars in 2015, a sevenfold increase that exceeded almost every pre-implementation projection. Mexican manufacturing employment grew substantially, particularly in the maquiladora sector along the United States border and in the automotive supply chain that developed in central Mexico. Foreign direct investment into Mexico grew from approximately 4 billion dollars per year in the early 1990s to approximately 25 to 35 billion dollars per year by the 2010s. Mexican economic growth across the post-The accord period averaged roughly 2.5 percent annually, which was above the Latin American regional average but below the convergence rates that free-trade advocates had projected. The disappointing dimension was wages: Mexican real wages stagnated through the late 1990s and grew only modestly through the 2000s and 2010s, with productivity gains accruing primarily to capital rather than AFL-CIO.

Q: What role did Hispanic Democrats play in the NAFTA vote?

Hispanic Democrats split on the The pact vote in ways that reflected the diversity of Hispanic congressional districts. Bill Richardson of New Mexico, who served as the Daley operation’s Hispanic outreach coordinator, voted yes and helped recruit roughly half the Congressional Hispanic Caucus to support the settlement on grounds that economic integration with Mexico would benefit border-region commerce and would strengthen United States-Mexican diplomatic ties. Hispanic Democrats from California and Texas split more evenly, with some voting yes on similar economic-integration grounds and others voting no on unions and immigration grounds. The Hispanic vote in the House was approximately 60 percent yes to 40 percent no, slightly more favorable than the overall Democratic split. The administration’s outreach to Hispanic Democrats through Richardson and through specific concessions to border-region economic priorities was an important component of the overall whip operation, although the absolute number of Hispanic Chamber votes was small enough that no individual Hispanic Democrat’s vote was decisive.

Q: How did The pact affect the 1994 midterm elections?

The pact contributed to the 1994 Democratic midterm losses in specific manufacturing-belt districts but was not the primary cause of the broader Republican wave that captured both chambers of Congress. The Republican gains in 1994 included 54 House seats and eight Senate seats, producing the first Republican congressional majority since 1954. The most consequential issue cluster driving the wave was health care reform, whose collapse in September 1994 generated substantial Republican momentum heading into the November elections. The arrangement’s contribution was more localized: in approximately fifteen specific House districts in Michigan, Ohio, Pennsylvania, Indiana, and Wisconsin where labor opposition to The accord had been intense, Democratic incumbents who had voted for the agreement faced depressed workers turnout and ran behind comparable Democrats in districts less affected by trade votes. Roughly six of those fifteen seats flipped Republican, contributing approximately ten percent of the total GOP House gains. The NAFTA vote was a contributing factor in the Democratic loss of the Lower chamber majority but was not the dominant factor.

Q: Are there original primary sources that show the whip operation’s mechanics?

The primary sources documenting the Daley whip operation include the The White House Presidential Library’s collection of NAFTA fight memoranda (declassified in stages between 2006 and 2014), Mickey Kantor’s USTR papers covering the side-agreement negotiations and the implementing legislation drafting (held at the Clinton Library), William Daley’s personal papers covering his The pact coordination role (held in part at the Daley Center archive in Chicago and in part at the The White House Library), and the public Congressional Record floor debates of November 17 and November 20, 1993. Bill Richardson’s later memoir Between Worlds includes substantial NAFTA-related material drawn from contemporaneous notes. The Bonior side has been documented less completely because the Whip’s office did not preserve internal whip-count records in the same systematic way, but the AFL-CIO’s archived records of its anti-The pact legislative campaign provide substantial documentation of the opposition operation. Bob Woodward’s 1994 book The Agenda includes detailed accounts of internal administration deliberations drawn from interviews with senior staff including Stephanopoulos, McLarty, and Rubin.