Hello. Welcome to this lecture on the relationship between unemployment and inflation, and the Phillips curve. In this class, we will be discussing the relationship between wages, unemployment, and inflation. For that, we will introduce the concept of the Phillips curve. When economists, policymakers, or the press talks about inflation, they usually refer to consumer price inflation, that is, how prices for bundle of consumer goods have changed. However, you may wonder what can cause consumer prices to go up. One answer is, input prices. Surely, you have noticed that filling up your car is more expensive when oil prices are high. The graph shows consumer and producer price indices. As you can see, while the producer price index is more volatile, the two series track each other really well. What is the major cost component of most industries? Labor cost. Let's add an employment cost index to this graph. You can see again that employment cost and consumer price inflation track each other really well. Moreover, consumer price inflation is often just between the producer price index and the employment cost index that includes wages and benefits, as it is the combination of the two. Now consider, when do wages increase most? When employees have the most bargaining power, that is, when unemployment is low, and there are less applicants per job. Conversely, when unemployment is high, employers have more choice, and they don't have to pay higher wages to attract job applicants. This relationship was first noted by Arthur Phillips, who plotted unemployment against the wage growth, and found an inverse relationship between unemployment and wage growth in the United Kingdom. Let's consider this more formally. The rate of wage inflation g_w, is the change in wages between time t, and t plus 1, divided by wages in t. The first factor for wage inflation is the unemployment gap, that is the difference between unemployment, u, and the natural rate of unemployment, u-star. The natural rate of unemployment means, that there is only frictional unemployment, that is, job search always takes time. There's always some unemployment even in a time of full employment. The unemployment gap therefore measures cyclical unemployment. The second factor for wage inflation is how responsive wage inflation is to this gap. Let's call this responsiveness Alpha. Putting these pieces together, we get wage inflation is the negative of the responsiveness of wage inflation to the unemployment gap times the unemployment gap. This equation is referred to as the simple Phillips curve. However, we have seen earlier that consumer prices and wages move together. Over time, the term Phillips curve has been used to describe the relationship between unemployment and inflation. Let's have a look at this relationship using data from the 1960s. As you can see, the negative relationship between inflation and unemployment holds in the United States in the 1960s. This graph suggests a clear trade-off between unemployment and inflation, and that trade-off is central to monetary policy as stable prices and maximum employment are the Federal Reserve's mandate. In the 1960s and early 1970s, it was thought that the relationship between unemployment and inflation was stable, at least in the long run. Policymakers therefore carefully explored this trade-off and used it for policy decisions. However, over the next two decades, a different pattern emerged. As the graph suggests, over the 1970s and 1980s, there does not appear to be a smooth relationship between inflation and unemployment in the long run. However, if you squint a little, you can still see a distinct pattern. It looks like there is not one, but three relationships. One for the early 1970s, one for the late 1970s, and one for the early 1980s. When you connect the dots, it looks like three Phillips curves, with the curve shifting outward over time. An outward shift is not a good thing. The further the Phillips curve shifts out, the more inflation is needed to reduce unemployment. While the 1970s and 1980s cast doubt on the long run relationship between inflation and unemployment, the data indicate that at least in the short run the relationship is present. We now examine what determines these short run Phillips curves in the 1970s and 1980s. Let's look at consumer price inflation between 1960 and 1995. Clearly, during the early 1970s and early 1980s, increases in consumer price inflation were on average much higher than in the 1960s. Hence, one possible explanation for the outward shift of the Phillips curve could be an upward shift in inflation expectations from the early 1970s to the early 1990s. To illustrate how inflation expectation matter in this context, consider your employer announces a five percent across the board wage increase. This sounds like a decent increase. But now suppose consumer price inflation currently is 10 percent and is expected to remain at 10 percent, so your income in real terms, that is income adjusted for inflation, actually falls by five percent. In other words, workers don't care as much about nominal wage increases as they do about real wage increases as loss of raising the standard of living. Let's account for inflation expectations in the Phillips curve. The expected real wage growth is the nominal wage increase, g_w, minus expected inflation, Pi^e. Plugging this in the simple Phillips curve gives that the expected real wage growth is equal to the negative of the responsiveness of wage inflation to the unemployment gap times the unemployment gap. If the real wage is constant, then nominal wage growth, g_w and consumer price inflation, Pi have to grow at the same pace. Replacing nominal wage growth with inflation gives us the inflation expectation-augmented Phillips curve that relates differences in inflation from inflation expectations to employment gaps. Let's rearrange this equation. From this formula, we can see that for full employment, meaning an unemployment gap of zero, inflation has to be equal to inflation expectations. Going back to the 1970s and 1980s, when workers expected higher inflation, many demanded higher wages pushing inflation up to the level of inflation expectations. This increase in inflation expectations account for the outward shift of the short-run Phillips curve. This is why policymakers often consider whether inflation expectations have changed. To be clear, other factors can also shift this Phillips curve. For instance, supply shocks, such as the oil shocks in the 1970s and 1980s. More advanced versions of the Phillips curve take also supply shocks into account. Again, the trade-off between inflation and unemployment is central for deciding on monetary policy. For instance, in 2008, the then vice chairman of the board of governors of the Federal Reserve System, Donald L. Kohn stated that an economic model of inflation is an indispensable input to monetary policy deliberations. A model in the Phillips curve tradition remains at the core of how most academic researchers and policymakers, including this one, think about fluctuations in inflation. Indeed, alternative frameworks seemed to lack solid economic foundations and empirical support. What have we discussed in this lesson? First, riches and inflation are moving together. Second, there is a trade-off between inflation and unemployment modeled in the Phillips curve. Third, changes in inflation expectations can shift the Phillips curve.
