Macroeconomics › Section M · The neoclassical perspective › Topic M.3
Balancing Keynesian and neoclassical models
Keynesian and neoclassical economics are often presented as rival camps, but most working economists use both. The skill is knowing which model suits which question, and where neither gives a complete answer.
Two maps, two scales
A street map and a road atlas of a whole country show the same land, but you would not plan a cross-country drive with one or find a café with the other. Macroeconomics has a similar pair. The Keynesian model, built on aggregate demand and sticky wages and prices, is a close-up map of the next few years. The neoclassical model, built on potential GDP and flexible prices, is a large-scale map of decades. Each is accurate for its own purpose and misleading if stretched too far. This topic sets out what each does well, where each struggles, and how modern economists combine them.
Keynesian strengths and limits
The Keynesian model is good at accounting for why recessions happen and why cyclical unemployment rises and falls. A collapse in spending, combined with wages and prices that adjust slowly, explains how an economy with intact factories and willing workers can produce far below its capacity. It also explains why spending shocks spread through the multiplier. Its weakness is the long run. By concentrating on short-run demand, it can overlook what drives long-term growth, and it says little about the natural rate of unemployment that persists even when the economy is at potential GDP.
Neoclassical strengths and limits
Neoclassical analysis starts from the supply side, asking what ultimately sets how much an economy can produce and how many people it can employ: capital, skills, technology and the workings of labour markets. It explains why living standards rise over decades and why some unemployment remains even in good times. Its weakness is the short run. It has little to say about why unemployment jumps over a year or two, and it is least convincing during a deep, long slump such as the Great Depression, when waiting for wages and prices to adjust meant years of mass unemployment.
Two views of inflation
The schools also weigh inflation differently. In the Keynesian view, with a downward-sloping short-run Phillips curve, a little extra inflation can be a price worth paying for lower unemployment, at least for a time. In the neoclassical view, with a vertical long-run Phillips curve, inflation buys no lasting reduction in unemployment, so it is a pure cost: it distorts price signals, redistributes wealth unpredictably and complicates planning. Both schools agree, however, that restraining aggregate demand is an effective way to bring inflation down.
The schools side by side
Keynesian analysis focuses on the short term; neoclassical on the long term. Keynesians stress sticky wages and prices; neoclassicals stress flexibility over time. In Keynesian analysis aggregate demand mainly determines output, and supply slopes upward; in neoclassical analysis aggregate supply determines output, and the long-run supply curve is vertical. The Keynesian Phillips curve slopes down; the neoclassical long-run curve is vertical. To reduce unemployment, Keynesians would raise demand to remove cyclical unemployment, while neoclassicals would reform labour-market institutions to lower the natural rate. On ending recessions, Keynesians see demand as an effective tool; neoclassicals see it as temporary at best, and possibly just a source of inflation.
Most economists use both
Macroeconomics is not a contest between pure Keynesians and pure neoclassicals. Robert Solow, who won the 1987 Nobel prize for his work on growth, described a common working approach: something like the Keynesian model for the short run, the neoclassical model for the very long run, and a patched-together hybrid for the awkward middle span of five to ten years. Modern research tries to build that hybrid. New Keynesian models combine forward-looking expectations with sticky prices, so demand matters in the short run while prices and supply adjust in the long run; central banks use such models. Real business cycle models, at the neoclassical end, explain fluctuations mainly through supply shocks. No single model has settled the debate.
Worked example: choosing the right model
Which framework fits each question best? 1. Why did unemployment jump during last year's recession? A short-run question about demand and sticky wages: start with the Keynesian model. 2. Why is output per person far higher today than in 1950? A question about capital, skills and technology over decades: the neoclassical model. 3. Will a tax cut lift output next year? Keynesian analysis predicts a short-run boost, but neoclassical analysis asks whether the economy is already near potential. 4. Will a five-year rise in deficits slow growth? A middle-span question: a hybrid view, combining demand effects with long-run effects on investment, is needed. 5. Check: the time horizon of the question decides the starting model.
Worked example: the Great Recession
The US recession of December 2007 to June 2009 tested both views. 1. Keynesian reading: falling house prices and a financial crisis cut consumption and investment, unemployment reached 10.0% in October 2009, and the cure was more demand. Congress passed a stimulus law costed at about $787 billion in 2009, and CBO later raised the estimate to roughly $830 billion. 2. Keynesian evidence: studies such as one by Alan Blinder and Mark Zandi concluded that fiscal and financial policies made the slump much milder. 3. Neoclassical concerns: lags, higher public debt, and a slow recovery that suggested the economy's potential had been damaged. 4. Conclusion: both lenses explained part of what happened, and estimates of the stimulus's effect still vary.
The 2020s: lessons for both sides
The pandemic recession of 2020 lasted only two months, from February to April, and the fiscal response was fast and large, weakening the old neoclassical objection that stimulus usually arrives too late. Then inflation surged, reaching 9.1% in June 2022. Some economists, including Lawrence Summers, himself usually counted a Keynesian, had warned in early 2021 that further stimulus risked pushing demand beyond potential. Others stressed supply disruptions and energy prices. Most analysts now see a mix of strong demand and constrained supply. The episode also revived a neoclassical point: people leaving the labour force and missed schooling can lower potential GDP itself.
Common mistakes to avoid
First, treating the two schools as one right and one wrong; each is strongest at a different time horizon. Second, thinking neoclassical economists deny that recessions happen; they argue recessions fade as prices adjust. Third, thinking Keynesians ignore inflation; they treat too much demand as its cause. Fourth, forgetting where the schools agree: both see restraining aggregate demand as a way to reduce inflation. Fifth, presenting any one recession as proof of a whole school; real episodes usually mix demand and supply forces.
Key takeaways
The Keynesian model explains short-run fluctuations and cyclical unemployment well but says little about long-run growth. The neoclassical model explains long-run growth and the natural rate well but struggles with short, sharp downturns and deep slumps. Keynesians see some inflation as a possible price for lower unemployment; neoclassicals see it as a pure cost. Many economists, following Solow's description, use Keynesian ideas for the short run, neoclassical ideas for the long run, and hybrid models such as New Keynesian models in between.
Flashcards
Flashcard
Explaining recessions and cyclical unemployment
Test your understanding
Question 1 of 5
An analyst wants to explain why US output per person is far higher today than in 1950. Which framework is the best starting point?
Question 2 of 5
On which point do Keynesian and neoclassical economists broadly agree?
Question 3 of 5
How does the neoclassical perspective view inflation, compared with the Keynesian perspective?
Question 4 of 5
What did Robert Solow suggest about using the two approaches?
Question 5 of 5
Which feature defines New Keynesian models?
