I. When Expansion Meets Constraint

The 1937 Court-packing crisis occurred at the precise peak of Roosevelt's political power — 523 electoral votes to 8 in 1936, Democrats holding 334 House seats to Republicans' 88 on entering the Seventy-Fifth Congress. A president at that height of authority, with that level of legislative advantage, still failed to restructure the Supreme Court. The failure is the most important fact about 1937.

This failure illustrates something that runs through this series as a structural constant: the system's checks are a living force. They do not wait for a weak president to activate. Here they emerged from Congress, from public opinion, and most significantly from within Roosevelt's own party. The Senate opposition to Court-packing was led not by Republicans but by a liberal Democrat, Burton Wheeler of Montana. No external enemy stopped Roosevelt; the system's internal structure did.

II. Background and Trigger

The immediate backdrop was the Supreme Court's destruction of several major New Deal programs. Schechter (1935) had struck down the NRA; Butler (1936) had invalidated the original AAA. By early 1937, the Wagner Act and Social Security faced serious threats. For Roosevelt, this was no longer a question of individual statutes but of whether the entire New Deal project could survive constitutional review.

The Court's political alignment was conventionally described as four conservative "Four Horsemen" against three liberal "Three Musketeers," with Chief Justice Hughes and Justice Roberts as swing votes. Roosevelt's own framing in his March 9 Fireside Chat was unusually direct: he accused the Court of having become "a third house of Congress — a super-legislature," arguing that the nation needed to act to "save the Constitution from the Court and the Court from itself." This framing positioned the argument not as an attack on judicial review in principle but as a defense of democratic self-governance against judicial overreach. Whether one finds the argument persuasive depends partly on what one thinks the Court was actually doing — a question on which historians still disagree. What is not in dispute is that the conflict was genuine and the stakes were real.

III. The Plan and Its Cover Story

The plan Roosevelt submitted on February 5, 1937 proposed that for every federal judge seventy or older who did not retire within six months, the president could appoint an additional judge, with the Supreme Court's maximum size set at fifteen. The public justification emphasized judicial efficiency and case backlogs.

Chief Justice Hughes destroyed this justification with a public letter to the Senate Judiciary Committee: the Court was fully current on its docket; increasing the number of justices would not improve but likely impede deliberation. Once the efficiency argument collapsed, the debate was exposed as what it actually was: a structural contest between the elected branches and judicial independence. The Senate Judiciary Committee's hostile report called the bill "unnecessary, ineffective, and extremely dangerous." Senators who had supported the New Deal's policies could now oppose the method with principled consistency — and they did.

The plan touched the system's deepest nerve because it proposed to alter an independent institution's composition to change its decisions. If the precedent had held, any future president commanding a congressional majority could pack the Court with compliant justices whenever a ruling went the wrong way. The danger was not primarily about the New Deal; it was about the permanent availability of this instrument to any future authority that found the Court inconvenient.

IV. Counterattack and Defeat

The opposition was broad and internally differentiated. In the House, Judiciary Committee Chairman Hatton Sumners blocked the bill's progress. In the Senate, Wheeler — a self-identified liberal who had supported New Deal programs — led the opposition. Modern research is direct on this point: the leader of the Senate effort against the bill was a liberal Democrat, not a conservative Republican.

This fact deserves sustained attention because it challenges the assumption that checks on power are tools of conservatism against reform. Some of the New Deal's firmest supporters helped stop Court-packing. They supported Roosevelt's economic program; they refused to pay for its constitutional survival by destroying judicial independence. The capacity to separate means from ends — to hold that a good objective does not justify any instrument — is a structural prerequisite for effective checks. If everyone always reduces institutional commitments to immediate policy preferences, the checks collapse into partisan alignment. 1937 showed this capacity still existed, at least for some.

The Senate Judiciary Committee voted an adverse report in June; the Senate voted 70 to 20 on July 22 to recommit the diluted version, effectively ending Court expansion. What remained was modest judicial administrative reform. Most assessments consider this Roosevelt's most significant political defeat of the second term. A president who had won 523 electoral votes a year earlier had, at the height of his power, run into the wall.

V. The Court's Pivot

Two Supreme Court decisions announced during the fight transformed the political landscape. West Coast Hotel v. Parrish (March 29, 1937) upheld Washington State's minimum wage law, effectively abandoning the rigid version of freedom-of-contract doctrine. NLRB v. Jones and Laughlin Steel Corp. (April 12, 1937) sustained the Wagner Act, holding that Congress could regulate activities within states having a close and substantial relation to interstate commerce.

The famous phrase "the switch in time that saved nine" has been substantially complicated by modern research. The Parrish conference vote took place on December 19, 1936 — before Roosevelt announced his plan on February 5, 1937. Roberts had already cast the decisive fifth vote before the packing plan was public. The evidence does not support the simple story of the Court caving to presidential threat. What drove Roberts is genuinely uncertain: some scholars emphasize legal-doctrinal factors; others see a response to the broader political environment of 1936; quantitative research identifies a real break in Roberts's voting pattern but explicitly rejects the naive version of the switch story. The most defensible conclusion is that the simple "threat caused capitulation" narrative is wrong, while leaving open whether political environment mattered at all.

The resolution of the crisis followed a pattern this series has seen before: not one side crushing the other, but the threatened institution adjusting its interpretation to accommodate what had grown around it while preserving its formal independence. The Court did not become a presidential instrument; it adapted without being captured. Expansion found the constitutional space it needed without destroying the check that had contested it — a characteristic resolution for this system, which tends toward mutual adjustment rather than one-sided dominance.

VI. Recession, Purge, and Presidential Limits

The 1937-38 recession — among the most severe contractions of the twentieth century — inflicted additional constraint. Real GDP fell roughly 10 percent; unemployment returned toward 20 percent; industrial production dropped 32 percent. The cause was not simply fiscal retrenchment but a combination: increased reserve requirements, Treasury gold sterilization since 1936, and the bite of new Social Security payroll taxes reducing aggregate demand. The lesson drawn by key administration officials — that premature stimulus withdrawal had extended the depression — became one of the New Deal's lasting macroeconomic contributions.

Roosevelt's 1938 "purge" attempt — open intervention in Democratic primaries to unseat conservative party members — failed almost completely. He targeted ten conservative Democrats; only one was defeated. The others returned to Washington more hostile. The implication: even a president commanding enormous electoral mandates cannot simply treat his party as an administrative extension. The party has its own institutional logic, its own constituent pressures, its own independent actors.

The 1938 midterms registered further losses — Democratic House seats fell from 334 to 262 — and helped consolidate the Conservative Coalition of Southern Democrats and Republicans that would constrain liberal legislation for decades. The internal contradiction of the New Deal coalition — its dependence on Southern Democratic votes to pass legislation, combined with Southern Democrats' resistance to anything threatening racial order — now hardened into a structural obstacle that would outlast the New Deal era itself.

VII. The Administrative State Keeps Growing

Court-packing failed; administrative expansion continued and deepened. The Brownlow Committee's 1937 finding — "the President needs help" — led to the Reorganization Act of 1939 and Reorganization Plan No. 1, which created the Executive Office of the President and moved the Bureau of the Budget and other management instruments directly under presidential control. Roosevelt told Congress explicitly: the president could not directly manage thirty to forty major agencies; reorganization was necessary.

The historical irony is precise. Court-packing failed; institutionalized presidential management capacity succeeded. This reveals a critical feature of the system's checks: they are not uniformly distributed. The defense of judicial independence was fierce, rapid, and cross-partisan — because the threat to it was visible, legible, and confronted an institution with the capacity to defend itself and mobilize others. But the expansion of presidential administrative capacity moved quietly, within the executive branch, without directly threatening any other institution's independence. The system had no comparably strong immune response to this kind of expansion.

The asymmetry matters: the same checks that successfully repelled the Court-packing scheme were nearly powerless against the quieter aggregation of administrative authority within the executive. A system that can forcefully block an overt takeover of an independent institution may still be unable to prevent the gradual, internal concentration of executive power. And it was precisely this more quietly assembled apparatus — now better organized and more institutionalized — that was about to encounter the largest external event of the century.

VIII. Conclusion

The 1937 crisis demonstrated that expansion meets real constraint within this system — and that the constraint can emerge from within the dominant party, from public opinion, and from institutional self-defense, not only from the opposing party.

The defeat was genuine. Roosevelt at the height of his power tried to alter the institutional conditions of judicial review and was stopped by his own senators, by the Chief Justice's public intervention, and by a public that had not resolved to support him. The checks were activated not by a weak president facing a hostile majority but by a powerful president facing a coalition mobilized by structural principle. The Court's own pivot — achieved without surrendering its independence — produced the characteristic resolution: mutual adjustment rather than one-sided victory.

But the check was partial and asymmetric. It held at the most visible point — the direct attempt to capture an independent institution — but was largely ineffective against the quieter administrative consolidation that followed. Court-packing was defeated; the Executive Office of the President was created. The system blocked the aggressive form while absorbing the incremental one.

Expansion and constraint, in this system, are not heading toward a final resting point where executive power is definitively limited or definitively unchecked. They are permanently in tension, each shaping the other, each temporarily winning one exchange while conceding another. 1937 held judicial independence; it let administrative organization proceed. The next round — total war — would give the reorganized, better-equipped administrative apparatus the largest call on its capacities it had ever faced, and would temporarily suspend many of the normal constraints in the name of emergency. That story belongs to the next essay.