Old Keynesian

The Old Keynesian tradition is the macroeconomic synthesis that emerged through the late 1930s and 1940s as Hicks, Hansen, Modigliani, Samuelson, and Tobin worked Keynes’s General Theory into a tractable framework that could sit alongside neoclassical microeconomics in graduate teaching and policy advice. By the early 1950s it had become the operating mainstream of macroeconomics in the English-speaking world, supplying the policy intuition behind the Kennedy-Johnson tax cuts, the assumption of a stable Phillips-Curve policy menu, and the routine practice of fiscal-and-monetary fine-tuning. By the late 1970s its reputation had broken against stagflation, and its substantive content was substantially absorbed into the New Keynesian successor program — but on terms set by Friedman, Lucas, and Phelps rather than by its own architects.

Thesis

Keynes’s General Theory (1936) demonstrated that capitalist economies can stall in equilibria below full employment and that there is no automatic mechanism returning them to full-employment equilibrium. The interest rate is a monetary phenomenon, not a real one, and adjusts to balance the demand for money against the supply rather than savings against investment. Aggregate demand is therefore the proximate determinant of output and employment over the cycle; investment is volatile and depends on long-period profit expectations under uncertainty; consumption depends on disposable income with a stable marginal propensity. The Phillips Curve (Phillips 1958, in the Samuelson-Solow 1960 reinterpretation) shows a stable empirical trade-off between inflation and unemployment that policymakers can read as a menu — a society chooses a position on the curve based on its preferences over inflation and unemployment outcomes. Fiscal and monetary policy can routinely shift the economy along this menu and toward full-employment equilibrium; periodic recalibration of demand-management settings is the routine business of macroeconomic policy.

Lead proponents

• John Maynard Keynes — the founding figure. The General Theory of Employment, Interest and Money (1936) is the foundational text; its arguments were refined and codified by Hicks, Hansen, and Modigliani over the following decade.

• John Hicks — Mr. Keynes and the Classics (1937) introduced the IS-LM diagram that became the canonical graphical statement of the Keynesian system. Hicks himself later (1980, “IS-LM: An Explanation”) expressed reservations about the apparatus he had originated; the textbook tradition retained it nonetheless.

• Alvin Hansen — Harvard economist whose A Guide to Keynes (1953) was the standard graduate-level Keynesian text in the 1950s. The principal channel by which Keynes’s thought entered American academic and policy circles.

• Paul Samuelson — Foundations of Economic Analysis (1947) gave neoclassical microeconomics the formal-mathematical apparatus it had previously lacked; Economics (1948, multiple editions) was the dominant US principles textbook for two generations and the principal vector for Old Keynesian doctrine in mass economic education. With Robert Solow, the 1960 paper introduced the Phillips Curve into US policy discourse as a stable trade-off.

• Robert Solow — neoclassical growth theorist whose 1956 paper supplied the canonical exogenous-savings growth model. Co-author with Samuelson on the 1960 Phillips Curve paper. Long career at MIT; Nobel 1987.

• Franco Modigliani — life-cycle hypothesis of consumption (1954); applied Keynesian econometrics; principal architect of large-scale econometric Keynesian forecasting models in the 1960s. Nobel 1985.

• James Tobin — the most articulate post-1960s defender of the Old Keynesian framework against monetarist and rational-expectations challenges. Asset Accumulation and Economic Activity (1980) and the 1972 AEA presidential address are the principal late-period statements. Nobel 1981.

• Lawrence Klein — the principal builder of the large-scale econometric Keynesian forecasting models that operationalised demand-management policy through the 1960s and 1970s; their failure on the 1970s data was one input to the methodological crisis the Lucas critique was responding to.

Key arguments

1. Aggregate demand is what binds output and employment. Capitalist economies can sustain underemployment equilibria; there is no automatic adjustment back to full employment. The proximate determinants of output are consumption (a function of income), investment (volatile, expectations-driven), government spending, and net exports. Stabilisation policy works by managing aggregate demand.

2. The Phillips Curve as a policy menu. Phillips’s 1958 empirical observation, in the Samuelson-Solow 1960 reinterpretation, identified a stable trade-off between inflation and unemployment. The trade-off was treated as exploitable: a society could choose a position with lower unemployment by accepting somewhat higher inflation, and policy should reflect that choice. This was the most consequentially mistaken claim of the Old Keynesian program; it pre-positioned the tradition for the stagflation reckoning.

3. The fiscal multiplier matters and is stable. Government spending raises GDP by more than its first-round impact through induced consumption; the multiplier is reasonably stable across normal conditions and ought to be the centrepiece of demand management. The 1964 Kennedy-Johnson tax cut was the canonical Old Keynesian policy success and was widely taken to confirm this position.

4. Money matters but mainly through interest rates. The IS-LM apparatus integrated the goods market and money market in a way that gave monetary policy a real but secondary role; “fiscalist” Old Keynesians tended to read it as showing fiscal policy was the more reliable instrument. The dispute with monetarism over the relative effectiveness of fiscal and monetary policy was conducted largely within the IS-LM framework.

5. Demand-management is routine policy. Stabilisation policy is the routine business of macroeconomic management, not an exceptional intervention. The central bank and Treasury should continuously calibrate fiscal and monetary settings to keep the economy near full employment with stable prices, exploiting the Phillips Curve trade-off where it is favourable.

Key evidence

• The 1960s expansion. The 1961–1969 expansion was the longest in US postwar history to that point, with low and stable inflation through 1965 and a sustained move down the Phillips Curve through the second half of the decade. Old Keynesian policy was widely credited; Walter Heller’s CEA chairmanship and the 1964 tax cut became canonical examples of the framework in action.

• The Bretton Woods era. The fixed-exchange-rate gold-dollar system (1944–1971) supplied a monetary anchor that simplified the operating environment for which the Old Keynesian framework was designed; the system’s collapse in 1971 removed that anchor and exposed framework limitations the prior environment had concealed.

• The Phillips Curve, 1958–1969. The US data through the late 1960s did show something close to a stable Phillips trade-off; the 1960 Samuelson-Solow paper and subsequent econometric work reflected the data accurately for that period. The framework’s failure was not in describing the period it was designed to fit but in mistaking a regime-conditional regularity for a structural one.

Major critiques

• The natural-rate hypothesis (Friedman 1968, Phelps 1967). Friedman and Phelps argued ahead of the data that the Old Keynesian Phillips Curve was short-run and conditional on adaptive expectations; sustained attempts to hold unemployment below the natural rate would produce accelerating inflation. The 1970s broadly confirmed this prediction. The Old Keynesian tradition’s most consequential single defeat.

• The Lucas critique (1976). Lucas argued that the econometric Phillips-Curve estimates the Old Keynesian tradition relied on were not just wrong about the 1970s but methodologically incoherent: the estimated parameters were functions of the policy regime, and using them to evaluate a different regime was theoretically invalid. The critique foreclosed Old Keynesian econometric practice in graduate teaching within a decade.

• Stagflation as falsification. Inflation and unemployment rose together through the 1970s. The Old Keynesian framework could be modified to accommodate this — supply-shock interpretations, shifting Phillips Curves, induced-expectations dynamics — but the modifications came at the cost of the framework’s predictive sharpness, and the framework had not predicted the episode in advance.

• Microfoundations imperialism. Even within the Keynesian tradition, the post-1976 generation argued that aggregate behavioural equations not derivable from explicit optimising behaviour were not legitimate macroeconomic theory. The New Keynesian program was, on this reading, an attempt to rescue Old Keynesian conclusions on New Classical methodological foundations — preserving the substantive commitments while abandoning the methodological ones.

Status today

Old Keynesian economics is no longer a research program at top-tier graduate departments and has not been since the late 1980s. Its substantive policy intuitions — that aggregate demand matters, that fiscal policy is sometimes effective, that the central bank can stabilise output, that the Phillips Curve is informative if properly conditioned — survived in the New Keynesian successor program, but on rational-expectations-and-microfoundations terms set by the New Classical methodological revolution. The intermediate-textbook tradition (Mankiw, Krugman-Wells, Blanchard) preserves IS-LM as a teaching device, with caveats. The fiscal-policy reassessment after 2008 — that fiscal multipliers are higher at the zero lower bound, that fiscal policy is more effective than the pre-crisis consensus had suggested — has produced a partial Old Keynesian revival in policy advisory work without re-establishing it as an academic research program.