This month marks the 75th anniversary of the publication of Keynes’s The General Theory of Employment, Interest, and Money (Keynes 1936). The impact of the General Theory is unquestionable. It became the dominant paradigm through the 1960s and today’s policymakers still cling to many of the General Theory’s tenets.
Google Scholar shows more than 12,000 citations to the General Theory – more than double the combined citations to Robert Lucas’s “Expectations and the Neutrality of Money” and Kydland and Prescott’s “Time to Build and Aggregate Fluctuations” – papers that helped supplant the General Theory as the major macroeconomic paradigm and helped earn their authors Nobel Prizes.
And in response to the recent economic crisis, some have attempted to resuscitate the General Theory in academia (Gordon 2009) and in the public sphere, evidenced by a debate in the Economist of the motion “This house believes that we are all Keynesians now”.
In our recent paper “The General Theory of Employment, Interest, and Money After 75 Years: The Importance of Being in the Right Place at the Right Time” (Luzzetti and Ohanian 2010), we analyse the impact of the General Theory on the economics profession. We argue that the significant and long-lasting influence of the General Theory stems from the fact that Keynes was in the right place at the right time. In particular:
- Macroeconomic time series from the 1940s through the 1960s conformed qualitatively to patterns predicted by the General Theory (even though the driving forces behind the US economy at this time may have been very different than the forces stressed by the Keynesian model);
- Simultaneous equation econometric developments made around this time elevated the Keynesian model to a quantitative enterprise;
- And perhaps most important, the General Theory was published during the Great Depression when there was enormous demand for new ideas to understand chronic depression.
But ultimately, just as the General Theory became the dominant macroeconomic paradigm, its influence waned for the same reasons that it had supplanted early equilibrium theories – its inability to explain subsequent macroeconomic developments when new theories could.
Why was the General Theory
The General Theory had its genesis during the Great Depression, the persistence and depth of which seemed to defy standard explanations of the time. Those explanations were based on equilibrium reasoning that held that price adjustment would equilibrate supply and demand.
Keynes jumped on the apparent inconsistency of using equilibrium theory to analyse prolonged depression by reviewing Pigou’s analysis of the labour market (Pigou, 1933). Pigou’s framework featured a labour-leisure tradeoff that is embedded in many of today’s standard models.
As the Depression persisted for years in the UK and the US, it became increasingly difficult to reconcile chronically high unemployment with equilibrium theory that posited wage adjustments would reduce unemployment to normal levels. The General Theory was, in large measure, written in response to the inability of equilibrium theory to confront the Great Depression.
Furthermore, US macroeconomic time series following the publication of the General Theory appeared consistent with Keynes’s predictions. As government spending soared in the 1940s, rising from about 16% of GDP in 1939 to 48% of GDP in 1944, the unemployment rate plummeted from 17.2% to 1.2% (Margo 1993). This increased economists’ confidence in the Keynesian model, and the stable and prosperous economy of the 1950s and 1960s further solidified this confidence.
But perhaps the central factor behind the longevity of the General Theory was a series of breakthroughs in econometric methods that began in the 1940s. These methodological developments transformed the qualitative ideas of the General Theory into quantitative propositions. These breakthroughs included Haavelmo’s 1944 paper that integrated more formally probability theory with econometric methods, and other Cowles Commission classics on identification, estimation, and causal ordering.
These econometric developments formed the basis of the toolkit used to analyse business cycles following the General Theory both among university economists and policymakers. Throughout the 1960s, the economy continued to grow with remarkable stability, and for many observers, this stable prosperity was due in considerable part to the General Theory’s tenets.
The decline of the Keynesian model
By the early 1970s, macroeconomic time series threw up patterns at odds with the Keynesian model:
- Poor forecasting performance of Keynesian econometric models,
- Increasing recognition of supply-side factors as drivers of fluctuations, and
- The breakdown of the Phillips curve.
Moreover, theoretical developments identified theoretical weaknesses of the Keynesian model and provided new theories grounded in the language of optimisation and equilibrium for understanding business cycles.
Charles Nelson (1972) levelled the first significant empirical criticism of the Keynesian model by showing that low-order integrated autoregressive-moving average models produced lower mean square error forecasts than detailed Keynesian models that were the industry standard.
Nowhere were these empirical inconsistencies more evident than with the Phillips curve.
Figures 1 and 2 demonstrate significant changes in the relationship between unemployment and inflation. This relationship, which was the focus of Samuelson and Solow’s (1960) famous discussion of the Phillips curve, is negative, with a correlation of around -0.6, between 1959 and 1969. Some economists interpreted this as an exploitable trade-off in which unemployment could be permanently kept at a low level provided that there was at least some inflation. But as is evident from Figure 2, there is virtually no relationship between these variables after the 1960s, as the regression slope between the two variables is indistinguishable from zero.
Figure 1. CPI inflation vs unemployment rate (1959-1969)
Figure 2. CPI inflation vs unemployment rate (1970-2010)
In addition, economists began to understand that a substantial fraction of fluctuations, particularly those in the 1970s, arise from the supply side of the economy. Kydland and Prescott (1982) used developments in general equilibrium theory to formalise this view as the real business cycle programme. They showed that roughly two-thirds of postwar fluctuations could be accounted for by variations in total factor productivity, i.e. a supply-side rather than demand-side factor.
The General Theory also advocated the importance of “animal spirits” in investment decisions. In equilibrium macroeconomic models, optimal decisions about the allocation of resources between current consumption and investment are summarised by the Euler equation. Thus, examining whether the Euler equation holds empirically is a natural test of the importance of animal spirits. To analyse the quantitative significance of animal spirits on physical investment, we use historical US data to construct Euler equation residuals under the assumption of perfect foresight (Luzzetti and Ohanian 2010). According to the Keynesian view, these residuals should be large and volatile, representing shifting expectations. Moreover, residuals should be negative during depressions, representing pessimism about future returns to capital.
We find – in contrast to the Keynesian view – that the residuals are small and appear to be uncorrelated in postwar data. And although the residuals are larger during the Great Depression, they are positive, suggesting that investors were optimistic in expecting higher returns than those that materialised.
The Keynesian framework, as presented in large-scale econometric models, was not microfounded, and thus was sharply at odds with the developments in economic theory of the 1960s and 1970s. Muth (1961) and subsequent research developed the theory of modelling expectations in dynamic settings. Lucas and Prescott (1974) and Mehra and Prescott (1980) showed how to integrate infinite-dimensional economies with recursive methods, making it feasible to quantitatively assess fully microfounded dynamic stochastic equilibrium economies. This all came together in 1982 with the seminal Kydland-Prescott paper. This framework allowed equilibrium macroeconomics to address questions that previously were considered beyond its grasp, such as the causes of the Great Depression (Cole and Ohanian 2004 and Ohanian 2009) and the World War II economic boom (McGrattan and Ohanian 2010). And equilibrium models are generating very different answers for understanding these episodes.
The General Theory’s staying power among policymakers
There is no doubt that the General Theory was one of the major economic events of the 20th century, at least as important as the impact of Kydland and Prescott (1982), which in turn dispensed with Keynesian economics. But some ideas from 1936 persist.
The notion of an inflation-unemployment trade-off and aggregate demand management remain at central banks, and the Keynesian vision provides a well-established framework for carrying this vision on within the context of policies that tie central bank behaviour to the joint mandate of promoting both low unemployment and price stability. This makes it politically unimaginable for a central bank, faced with a crisis, to argue it is unlikely they can increase output and trying to do so might make matters worse.
The General Theory will continue to have a large audience among policymakers as long as governments are pressed to boost nominal spending during periods of crisis, whether or not those efforts are effective.
Cole, H and L Ohanian (2004), “New Deal Policies and the Persistence of the Great Depression: A General Equilibrium Analysis”, Journal of Political Economy, 112(4): 779-816.
Gordon, R (2009), “Is Modern Macro or 1978-era Macro More Relevant to the Understanding of the Current Economic Crisis?”, Unpublished Paper.
Haavelmo, T (1944), “The Probability Approach in Econometrics”, Econometrica 12, July, pp. iii-vi+1-115.
Keynes, J (1936), The General Theory of Employment, Interest, and Money, Harcourt, Brace & World.
Kydland, F and E Prescott (1982), “Time to Build and Aggregate Fluctuations”, Econometrica 50(6):1345-1370.
Lucas, R Jr. (1972), “Expectations and the Neutrality of Money”, Journal of Economic Theory, 4(2), April:103-124.
Lucas, R Jr. and E Prescott (1974), “Equilibrium Search and Unemployment”, Journal of Economic Theory, 7(2):188-209.
Luzzetti, M and L Ohanian (2010), “The General Theory of Employment, Interest, and Money After 75 Years: The Importance of Being in the Right Place at the Right Time”, NBER Working Paper 16631, December, forthcoming in Keynes’ General Theory: Seventy-Five Years later, Tom Cate, editor, Edward Elgar: London.
Margo, R (1993), “Employment and Unemployment in the 1930s”, Journal of Economic Perspectives, 7(2):41-59.
McGrattan, E and L Ohanian (2010), “Does Neoclassical Theory Account for the Effects of Big Fiscal Shocks? Evidence from World War II”, International Economic Review, forthcoming.
Mehra, R and E Prescott (1980), “Recursive Competitive Equilibrium: The Case of Homogeneous Households”, Econometrica, 48(6):1365-79.
Muth, J (1961), “Rational Expectations and the Theory of Price Movements”, Econometrica, 29(3):315-335.
Nelson, C (1972), “The Prediction Performance of the FRB-MIT-PENN Model of the US Economy”, American Economic Review, 62(5):902-917.
Ohanian, L (2009), “What - or who - started the great depression?”, Journal of Economic Theory, 144(6):2310-2335.
Pigou, A (1933), The Theory of Unemployment, Macmillan and Co., Ltd.
Samuelson, P and R Solow (1960), “Analytical Aspects of Anti-Inflation Policy”, American Economic Review, 50(2), Papers and Proceedings of the Seventy-second Annual Meeting of the American Economic Association,177-194.