The Neoclassical Perspective

This is Unit Ten of Principles of Macroeconomics: the neoclassical perspective. Unit nine gave the case for acting on demand. This unit gives the case for expecting the economy to right itself. Carry one idea through. The two perspectives are not rivals to be scored: they answer different questions on different horizons, and their whole disagreement reduces to one quantity, which is how long the middle of the story takes. The capstone tells a single contraction twice, once each way, and both accounts are internally consistent on the same diagram. By the end you will work the self-correction sequence in both directions, say which Phillips curve a claim describes, read a natural rate out of paired data, name which of the three policy lags a delay is, and say which perspective a question calls for.

Four entries on where the neoclassical account starts, and the curve that follows from it.

The starting point.

MAC 0901 The Neoclassical Starting Point

The neoclassical perspective reverses unit nine's emphasis. Over long periods, potential GDP determines the size of the economy, and aggregate demand determines only the price level. The classical tradition held this before Keynes, stressing flexible prices returning the economy to full employment. Neither view claims the other is always wrong: they claim different horizons.

The assumption doing the work.

MAC 0902 Wages and Prices Are Flexible Over Time

The second building block answers unit nine directly. Wages and prices are sticky in the short run but flexible over longer periods, so the economy returns to potential on its own: the stickiness that traps output below potential for a year wears off over several. Unit nine denied this. The disagreement is about how long, not whether.

What full employment means.

MAC 0903 Full Employment Is Not Zero Unemployment

Full employment means zero cyclical unemployment, not zero unemployment. Frictional and structural unemployment remain even at potential output, and together they are the natural rate. An economy at potential still has people between jobs and people whose skills no longer match. A policy promising zero unemployment is promising something the term does not mean.

The curve that follows.

MAC 0904 The Vertical Long-Run Supply Curve

Long-run aggregate supply is vertical at potential GDP. Aggregate demand cannot change long-run output; only potential GDP can, and the curve drifts right over time, which is growth. Physical capital per person, human capital and technology move it; spending does not. Vertical does not mean fixed: it means unresponsive to the price level.

Two entries working the sequence that returns output to potential, and the conclusion drawn from it.

The sequence, both ways.

MAC 0905 Working the Self-Correction Sequence

Trace a demand shift through both curves until the economy is back at potential. The sequence is always four steps: demand shifts, the short run lands off potential, the labour market responds, and short-run supply shifts until potential is restored. Take potential GDP of seven thousand billion. Demand shifts right, output rises above seven thousand and unemployment falls below the natural rate; employers compete for scarce workers, nominal wages rise, that is an input price increase, so short-run supply shifts left and output returns to seven thousand at a higher price level. Run it the other way: demand shifts left, output falls below potential, unemployment rises, wage growth stalls, supply shifts right, and output returns to seven thousand at a lower price level. Both directions end at the same output and differ only in the price level, which is the whole neoclassical claim. Note what the sequence does not say: it gives no date.

The conclusion drawn from it.

MAC 0906 Output Above Potential Cannot Last

The sequence has a one-line conclusion. An economy cannot sustain output above potential, because the labour market bids wages up and supply shifts back, so a demand surge ends as a higher price level rather than more output. The extra output is real while it lasts, and temporary. This says nothing about how long, which is where the schools part company.

Two entries on what people are assumed to know, and how long the adjustment takes.

What people are assumed to know.

MAC 0907 Rational and Adaptive Expectations

Two accounts of expectations. Adaptive expectations say people learn from past experience and adjust gradually. Rational expectations say people use all available information, including announced policy, and act on it at once, so an anticipated policy is priced in before it arrives. The two differ in speed, and therefore in whether policy can move real output.

How long the adjustment takes.

MAC 0908 How Long Is the Short Run

Nobody knows precisely. The working estimate is that the short-run effect of a demand shift lasts roughly two to five years, which makes Keynesian analysis suit shorter horizons and neoclassical analysis longer ones. Keynes answered that appeal with the remark that in the long run we are all dead. Treat the figure as an estimate under dispute, not a measurement.

Four entries on what this does to the Phillips curve, and what sets inflation over long periods.

What it does to the curve.

MAC 0909 The Long-Run Phillips Curve Is Vertical

If long-run supply is vertical at potential, unemployment sits at the natural rate whatever inflation does. So the long-run Phillips curve is vertical at the natural rate, and unit nine's tradeoff disappears. Friedman put it in nineteen sixty-seven as a temporary tradeoff always, a permanent tradeoff never. The short-run curve still slopes down, and both are true of different horizons.

Telling the two apart.

MAC 0910 Short Run or Long Run

Given a claim about inflation and unemployment, say which curve it describes. The test is whether expectations have finished adjusting. If people are still being surprised by the price level, the short-run curve governs and a tradeoff exists. If expectations have caught up and output has returned to potential, the long-run curve governs and no tradeoff remains. A claim that stimulus lowered unemployment this year while raising inflation describes a movement along the short-run curve. A claim that a decade of higher average inflation left average unemployment unchanged describes the long-run curve. A claim that policy can hold unemployment permanently below the natural rate by accepting steady inflation contradicts the long-run curve, and it is the claim the nineteen seventies tested and broke. The long-run curve does not say policy is powerless: it says demand policy cannot move unemployment permanently.

Finding that rate in data.

MAC 0911 Reading a Natural Rate from Data

Given paired inflation and unemployment figures over several periods, find the natural rate. The method follows from the vertical curve: if unemployment always returns to one value while inflation takes many values, that recurring value is the natural rate. Look for the rate the data keeps coming back to, not the average and not the lowest. Take six periods where unemployment runs five point four, four point one, five point four, six point eight, five point four and four point three percent, while inflation runs two, four and a half, five, three point two, seven and a half and nine percent. Unemployment returns to five point four three times, and at those moments inflation is two, five and seven and a half percent, which are far apart. One unemployment rate is consistent with three very different inflation rates, so the natural rate is five point four. The average of the six figures is about five point two, close enough to be mistaken for it, and that closeness is a coincidence of this table rather than a method.

What sets inflation over the long run.

MAC 0912 Inflation in the Long Run

Over long periods inflation is set by two rates: how fast money grows and how fast the economy grows. Money growth above real growth produces inflation, and below it, deflation. A study of a hundred and sixteen countries from nineteen sixty to two thousand and seven found the relationship strongest where inflation averaged above ten percent. This is a result; the equation it comes from is unit twelve's.

Three entries on what policy can still do, once demand policy is set aside.

Moving the rate itself.

MAC 0913 Lowering the Natural Rate

If demand policy cannot move long-run unemployment, the neoclassical answer is to move the natural rate instead, by acting on frictional and structural unemployment: job matching, information, training and benefit design. Lowering the natural rate raises potential GDP, because more people are working. It is a different instrument, on a different quantity, over years rather than quarters.

The first component.

MAC 0914 Frictional Unemployment and the Job Search

Frictional unemployment arises because matching takes time. A searcher holds a reservation wage, the lowest offer they would accept; it falls as the search runs on, the best offer received rises, and the search ends where the two meet. Cheaper job-market information shortens the search and lowers frictional unemployment. Unemployment compensation lengthens the search, raising both measured unemployment and the accepted wage.

The second component.

MAC 0915 Structural Unemployment and Mismatch

Structural unemployment exists when a firm is hiring and a worker is searching but the skills do not match. Technological change, shifts in what consumers buy, and regional differences produce it. Assembly work that once asked mainly for reliability and dexterity now tests analytical and communication skills. This is a mismatch of characteristics, not a shortage of jobs or of willingness.

Two entries on why acting at all is argued against, and how to tell the three delays apart.

The argument against acting.

MAC 0916 The Case Against Fine-Tuning

Policy takes time at three stages. Recognition, because output data arrive late and get revised. Implementation, because acting takes time. And impact, because the effect on demand arrives later still. A typical postwar recession ran about a year, which is short enough for help to arrive afterwards. The anchor calls the middle stage the legislative lag, and unit fourteen works all three in full.

Telling them apart.

MAC 0917 Name the Lag

Given a description of a delay, name which of the three lags it is. The test is what is being waited on. Recognition waits on information about where the economy is. Implementation waits on the decision and the machinery that carries it out. Impact waits on the economy itself to respond to a change already made. Output figures for a quarter arriving months later and then being revised is recognition. A tax change proposed, debated for a year and taking effect the following year is implementation. A rate cut taking effect immediately while borrowing and hiring respond over the following year is impact. Add the three and a response to a one-year recession can land after it has ended. They are cumulative, so naming one correctly matters less than adding all three: an argument that any single lag is short does not show the total is short.

One entry to close the unit: one contraction told twice, and what the schools actually disagree about.

Holding both at once.

MAC 0918 Two Horses, One Economy

Here the unit's governing idea is at work. The two perspectives are not rivals to be scored; they answer different questions, and the skill is knowing which one a question calls for. Keynesian analysis explains recessions and cyclical unemployment, and why a downturn can persist, but has little to say about growth or the natural rate. Neoclassical analysis explains growth, potential output and the structure of the labour market, and says the economy returns to potential on its own, but explains short-run swings poorly and deep prolonged recessions worst of all. Solow's formulation is that at short time scales something Keynesian is a good approximation, while at very long time scales the interesting questions are best studied in a neoclassical framework, and the anchor's image is a circus rider standing on two horses who needs both. Take one demand contraction, told twice. The Keynesian way: spending falls, the expenditure line drops, output settles below potential, unemployment is real and lasting, and the case for acting follows. The neoclassical way: the same contraction opens a recessionary gap, unemployment above the natural rate slows wage growth, short-run supply shifts right, and output returns to potential at a lower price level with no policy at all. Both accounts are internally consistent and both use the same diagram. They differ on one quantity — how long the middle of the story takes — and the policy disagreement follows entirely from that. Three cautions. The self-correction sequence is a claim rather than an observation, and it has a named challenger: efficiency-wage theory argues firms may deliberately hold wages above the market-clearing level to retain and motivate workers, in which case the wage adjustment that closes a recessionary gap never arrives and the gap becomes permanent. Yellen set that out in nineteen eighty-four, and whether efficiency wages operate at this scale remains disputed. Second, the schools were shaped by the episodes they had to explain, and an argument that fits one decade is not thereby general. Third, the profession has not resolved this: surveys find most economists working in the synthesis that treats prices as sticky and demand policy as usable, while most of those same economists also prefer that discretionary fiscal policy be avoided.

Look at what you can now do. You can work the self-correction sequence in both directions, through the labour market and the input price it changes, and say that it gives no date. You can say which Phillips curve a claim describes, by asking whether expectations have finished adjusting. You can read a natural rate out of paired data by finding the recurring value rather than the average. You can name which of the three lags a delay is, and remember they add up. And you can say which perspective a question calls for, knowing the two differ on one quantity rather than on the diagram. Unit eleven turns to where saving becomes someone else's borrowing.