EC 235 | Fall 2023
Required readings:
After studying the labor market and how it influences wage- and price-setting behavior, it is time to connect (un)employment and inflation.
Whenever the price level rises in a consistent and generalized way, we may say that an inflationary process is taking place.
And one of the ways in which we can study this phenomenon is through what happens in the labor market.
Recall the wage-setting relation:
\[ W = P^e \cdot F(u,z) \]
So far, we have not explicitly assumed a form to the function F.
Let us start by a linear form:
\[ F(u,z) = 1 - \alpha u + z \]
\[ F(u,z) = 1 - \alpha u + z \]
where α captures the strength of the effect of unemployment on wages.
Then, the price-setting relation becomes:
\[ W = P^e (1 - \alpha u + z) \]
And recall the price-setting relation:
\[ P = (1+m)W \]
Using the wage- and price-setting equations together:
\[ P = P^e(1+m)(1 - \alpha u + z) \]
… And this expression gives us a relation between the price level, the expected price level, and the unemployment rate.
In case we want to replace the price level by the inflation rate, π, and the expected inflation rate, πe, we end up with
\[ \pi = \pi^e + (m+z) - \alpha u \]
What are the effects of:
A rise in the markup rate, m?
An increase in expected inflation, πe?
A decrease in the unemployment rate, u?
For a better understanding of what is to come, it will be convenient to include time indexes in our previous inflation equation:
\[ \pi_t = \pi_t^e + (m+z) - \alpha u_t \]
where the subscript t refers to the present period; t-1 to the previous period, and so on.
In 1958, A. W. Phillips (1914–1975) empirically found that low rates of unemployment were associated with high rates of inflation, and high unemployment with low inflation.
This relationship has since been referred to as the Phillips curve.
From the Phillips curve equation:
\[ \pi_t = \pi_t^e + (m+z) - \alpha u_t \]
Assume, first, a stable scenario, with no persistent change in the price level.
This can be represented by πe = 0.
Then, it becomes
\[ \pi_t = (m+z) - \alpha u_t \]
But then…
Why did the original Phillips curve change?
We need to bring back πe.
Now,
\[ \pi_t^e = (1-\theta)\bar{\pi} + \theta \pi_{t-1} \]
When θ = 1, we have:
\[ \Delta \pi_t = (m + z) - \alpha u_t \]
where Δπt = πt - πt-1 is the change in the inflation rate.
High unemployment leads to decreasing inflation; low unemployment leads to increasing inflation.
This version is known as the accelerationist Phillips curve.
When we studied the labor market, we defined the natural rate of unemployment as the unemployment rate at which the actual price level (P) is equal to the expected price level (Pe).
Applying back this idea with our Phillips curve equation, we have π = πe:
\[ 0 = (m+z) - \alpha u \]
and the natural rate of unemployment (un) is given by
\[ u_n = \dfrac{(m+z)}{\alpha} \]
What does this relation imply for unemployment and inflation?
EC 235 - Prof. Santetti