Short-Run Aggregate Supply/Aggregate Demand and Policy

Abstract

This chapter deals with relationships between the price level and real output. A higher price level is linked with a reduced level of aggregate demand (AD), but with an increased level of aggregate supply (AS). The curves for AD and AS are first explained in conventional Keynesian terms, where a change in the price level causes a change in the quantity demanded and the quantity supplied; AD is shaped by the central bank’s response to inflation, while AS is driven by stories of sticky wages or worker misperceptions. These are followed by alternative explanations: AD is shaped by the uncertain transition from planned expenditure to achieved real demand. AS results from the varying degree of firms’ willingness and ability to respond to increased demand by increasing output. The chapter ends with a discussion of the Lucas critique and an extension of that insight to broader issues of how policy is perceived.

Appendices

Appendix: The Phillips Curve

In 1958, the economist A.W.H. Phillips published a paper looking at the connection between the rate of unemployment and the rate of growth of nominal wages [6]. As James Forder [2] documents, there was already a literature on the role of the unemployment rate in determining wages, but somehow the name “Phillips curve” got attached to it, and then used in a variety of subtly different meanings, including a common textbook version (e.g., DeLong [1]), in which it’s a relationship between unemployment and the rate of inflation.

This version of the Phillips Curve is anchored by expected inflation and the natural rate of unemployment.

Expected inflation is straightforward: the rate of inflation that people in the economy expect over the coming time, perhaps a year.

The natural rate of unemployment is also known in this context as the NAIRU, or the non-accelerating inflation rate of unemployment. The name is awkward, but the concept is fairly intuitive. If unemployment is very low, that’s a sign of there being little “slack” in the economy, few productive inputs that aren’t being put to use. The price of those inputs—particularly the wage to pay labor—will be pushed up, feeding inflation. As people observe actual inflation that’s higher than what they expected, eventually their perceptions will shift and they’ll start to expect higher inflation. Those expectations will get built into future price-setting behavior, making the expectation reality. If unemployment remains low, there will still be a lack of slack in the economy, and so prices will continue to get pushed up, but now from the higher base of people’s higher expectations, leading to continually rising inflation. The NAIRU is the lowest rate of unemployment that won’t cause this phenomenon. Above the NAIRU, inflation is stable or even falling; below it, you eventually trigger the continual rise described here.⁵

Figure 16.11 shows a representation of the Phillips curve (the downward-sloping line labeled “PC”), anchored at the expected-inflation rate π ^e and natural rate of unemployment u ^∗.

Fig. 16.11

A drop in aggregate demand

The Phillips curve can be seen as a kind of aggregate supply curve, just flipped around. In the normal aggregate supply curve, higher prices go with higher output; in the Phillips curve, higher inflation goes with lower unemployment, which implies higher output.

In this framework, if the Phillips curve is a type of aggregate supply curve, then the aggregate demand curve is something known as the monetary policy reaction function, or MPRF. The idea is that the monetary authority is concerned about inflation. At higher levels of inflation, it will be more concerned and so is more likely to raise the interest rates it controls. That in turn should slow down expenditure, leading to lower GDP (or lower growth of GDP) and the the higher unemployment associated with that. So the MPRF is an upward-sloping line, as shown in Fig. 16.11. Equilibrium in the economy is at their intersection (see point “A” in Fig. 16.11). An exogenous decrease in aggregate demand (say, a loss of consumer confidence) would cause the MPRF to shift to the right, moving equilibrium to point “B”.

A limitation of the Phillips curve is that its two anchor points—expected inflation and the NAIRU—are things we can’t actually observe in the world. We are instead forced to work with proxies for them or other ways of estimating them. Still, it can be a useful tool in thinking about expectations and how they influence economic outcomes, somewhat analogously to the role played by perceptions of how good policy is, described in Sect. 16.6.1. If inflation expectations are “anchored,” that implies that they don’t change too easily and policy makers can tolerate a modest period of somewhat elevated inflation without triggering a change in expectations and a movement of the Phillips curve in an unfortunate direction. In contrast, if inflation expectations are not well anchored, much more caution is in order, so as not to spook the public into expecting higher inflation, which would leave future policy-makers with less favorable tradeoffs.

Problems

Problem 16.1

• (a) Use the parameters in this table to solve for the IS and LM curves.

  Table 1

• (b) Use the IS and LM curves from (a) to solve for r and Y.

• (c) Now consider an increase in G from 400 to 425. Which curve moves? Which way does it move? How far does it move?

• (d) Using the new curve from (c), solve for the new values of r and Y. (e) How did the actual change in Y (from (b) to (d) differ from the size of the shift of the curve that you identified in (c)? Why is it different, and why is it different in the way it is (i.e., smaller or larger)?

• (f) Use the parameters below to solve for output Y and the price level P.

  Table 2

• (g) When G increases from 400 to 425, as in (c), which of these two curves (AD or AS) is affected, which way does it move, and how far does it move?

• (h) Use your new curve from (g) to calculate new values of Y and P.

• (i) How does your answer to (h) change either the IS or the LM curve?

• (j) Use the updated curve from (c) and the updated curve from (i) to solve for r and Y.

• (k) How does the change from (b) to (j) compare with the earlier changes (part (b) to part (d), and part (b) to part (e))? Why is the (b)-to-(j) change different in the way that it is?