Birks – Mankiw Chapter 35: The Short-run Trade-off Between Inflation and Unemployment
Mankiw, N. G. (2015) Principles of economics (7th ed.) Ch.35
Principles of macroeconomics (7th ed.) Ch.22
Mason, OH: South-Western Cengage Learning
The Short-run Trade-off Between Inflation and Unemployment
When reading the chapter, here are some aspects to consider:
- This chapter considers the original Phillips Curve and subsequent developments. There is a specific representation given, including links between an ADAS framework and a Phillips Curve framework, both in the short run and the long run. Other textbooks may present this in other ways.
- The story begins with the observation of an apparent relationship between inflation and unemployment. Once observed, it was considered to present policy options. When the relationship appeared to break down, the relationship was considered in more detail, with attempts to find possible theoretical structures which might explain both the original and later observations. In terms of “framing”, we can see how it was perhaps almost by chance that economists developed theories of inflation unemployment trade-offs. Once thinking along those lines, later developments occurred within that chosen context. We can anticipate further revisions to the thinking in time as other situations are observed, but we do now have possible explanations which are consistent with what has been observed in recent decades.
- It may be that some past observations related to a time when, for example, monetary policies were passive, accommodating changes in demand for money. It does not necessarily follow that the same relationships would be observed if monetary policy is actively used to change inflation or output levels.
- Mankiw is trying to link two models. Consider the two graphs. One has price on the vertical axis, while the other has inflation. This causes some complications, as you see with the explanation for Figure 2. It is not possible to read across directly from one to the other. Consequently the two graphs are not directly related as are the three in the model of Figure 4 in Chapter 32. Here you have to imagine figure 2(a) referring to two alternatives in some particular year, with the price levels of 102 and 106 representing possible alternative rises in the price level from the previous year, when the price level was 100. This suggests a pattern of annual revisions of the ADAS model as inflation occurs. This might lead us to imagine steady upward movements in both the AD and (short-run) AS curves as prices and costs rise with inflation. Mankiw does not tell you this. It is a “back story” that he hints at in the text in Section 1b when he refers to 2020 and 2021 (but he says nothing about any changes in AS). If you look up “Phillips Curve” in some of the longer introductory economics texts, you are likely to find an explanation which relates the short run Phillips curve to AD-AS curves which are moving upwards each year as the price level changes. Instead, Mankiw presents Phillips curves and AS curves (both short- and long-run) as if they resemble mirror images of each other, but with different variables on the axes.
- Note the similarities between the LR Phillips Curve and the LRAS curve. If LRAS is independent of the price level (an assumption of this theory), it is also independent of changes in the price level, which suggests that the labour market, and hence unemployment, is similarly independent.
- According to Mankiw’s story:
* the SR Phillips Curve is the result of AD shifting along a SR AS curve, and
* the LR Phillips Curve is the result of AD shifting along the LR AS curve
- Note the adjustment formulae at the start of Section 2d and earlier towards the end of Chapter 20 Section 4d. These are not general specifications. They give a specific linear relationship for the change in Y in response to the difference between expected and actual values of X.
- Figure 6 might look like a short-run Phillips Curve, but there is no guarantee that all the points are from the same curve. They are for eight different years. Assuming that the relationship is truly short run, they may be from several different curves as expectations and/or curves shift. This illustrates one of many problems in the interpretation of data.
- Mankiw also describes the sacrifice ratio. Note that this again is a linear relationship, which might not be realistic – think of reducing inflation from say 50% or 100% per annum. A 20% drop in inflation might not be unreasonable in either of these cases, but we would not expect output to fall to zero as suggested by a sacrifice ratio of 5.
Commentary by Stuart Birks, 8 September 2014