Phillips Curve - Learn How Employment and Inflation are Related
However, from the 's and 's onward, rates of inflation and unemployment differed from the Phillips curve's prediction. The relationship between the two. Graph of the short-term relationship between unemployment and inflation Samuelson and Solow named the relation after A.W. Phillips. Related Reading. Phillips found a consistent inverse relationship: when unemployment was high, wages increased Figure 1 indicates that the cost, in terms of higher inflation, would be a little more than half a percentage point. . Further Reading Phillips, A. W. H. “The Relation Between Unemployment and the Rate of Change of Money.
A brief review of each of these issues is provided. Before turning to the discussion, there are two factors that have likely played important roles in inflation developments in recent years that receive only a short discussion herein.
First, wage developments, and in particularly the interaction of low inflation and downward nominal wage rigidity, have likely played a role; Daly and Hobijn review related evidence. In addition, the appreciation of the dollar over the past year has been a factor in recent developments. Inflation in recent years has been low by the standards of the past half century, and the recent decline in oil and hence energy prices is not an important factor in explaining the low average pace of inflation in recent years, as the average pace of inflation since is essentially identical for the overall and core CPIs.
Percent change is the change over the previous twelve months. The data shown in the figure spans the period from January to September Figure 2 presents a scatterplot of the change in core CPI inflation against the unemployment rate, as suggested by this Phillips curve.
Over the period, a strong negative relationship between the level of unemployment jumps out of the figure and is captured in the fitted-regression line through the red dots.
In contrast, there is little relationship between the change in inflation and unemployment over the period the blue dots and fitted line. Indeed, the decline in inflation since is surprisingly small using the pre relationship, as can be seen through a comparison of the change in inflation when the unemployment rate is near 10 percent in the blue dots with the change that occurred in the early s recession in the red dots for unemployment near 10 percent.
Core inflation is measured by the percent change in the annual average of the CPI from its level in the previous year. Department of Labor, Bureau of Labor Statistics and author's calculations. An alternative is that current inflation could be anchored by relatively stable inflation expectations, with such expectations replacing lagged inflation in the relationship between inflation and unemployment.
This effect is predicted by New-Keynesian Phillips curves with a stable inflation target, as in Gali and GertlerKileyor the current generation of dynamic-stochastic-general equilibrium DSGE models used at central banks which build on, for example, Smets and Wouters Figure 3 explores the relationship between inflation expectations inflation one-year ahead, as measured in the first quarter of the year and the rate of CPI inflation in the previous year as measured by the percent change in the annual average of the CPI in the previous year.
Inflation and Inflation Expectations Note: Inflation in the previous year is measured by the percent change in the annual average of the CPI from the level in the previous year.
Inflation expectations 1-year ahead are measured by the one-year expected inflation in GDP prices for the period, as reported in the first-quarter Survey of Professional Forecasters.
For the period, inflation expectations are measured by the one-year expected inflation in the CPI from the Survey of Professional Forecasters.
Accessible version The relationship between expected inflation and lagged inflation weakened substantially moving from the pre period to post period the red to blue dots. As a result, an expectations-augmented Phillips curve would predict substantially less cumulative disinflation in response to persistently elevated unemployment, as the anchoring of inflation expectations limits the degree of disinflation.
Even so, this anchoring is probably insufficient to explain the limited decline in inflation following the rise in unemployment in and Returning to figure 2, both expected and actual core inflation were in the neighborhood of 2 percent during those years, and hence the fact that inflation remained near this level when unemployment was near 10 percent as can be seen by the small change in inflation when unemployment was near 10 percent in the blue dots points to factors other than anchored inflation expectations.
A second factor explored by Ball and Mazumderand emphasized by Del Negro, Giannoni, and Schorfheideis a "flat" Phillips curve. Three related interpretations of "flatness" in the Phillips curve have been offered. Ball and Mazumder suggest that "menu cost" models of nominal price and wage rigidity imply that such rigidities increase as the average rate of inflation falls, implying that more of the adjustment in nominal aggregate demand falls on output and less on inflation when inflation is low; this is exactly the finding emphasized in Kileywhich analyzed support for this prediction across a large sample of countries.
Research exploring the effects of downward nominal-wage rigidity points to a reduced effect of labor-market weakness on inflation in a low-inflation environment Daly and Hobijn In addition, Del Negro, Giannoni, and Schorfheide suggest that structural models of inflation dynamics in the New-Keynesian tradition have long suggested "flatter" Phillips curves e. Finally, Kiley and Boivin, Kiley, and Mishkin present evidence that a more clear commitment to price stability in recent decades, in the form of a monetary policy rule with a more sizable response to inflation, acts to substantially stabilize inflation expectations and mitigate fluctuations in inflation, with less clear effects on economic activity; such a shift in monetary policy behavior is consistent with an observed flattening in the Phillips curve.
While a flattening of the Phillips curve is consistent with a number of theoretical and empirical studies, the links between structural features of the economy and the properties of the reduced-form relationship between inflation and unemployment is complex and depends on many model features. Perhaps for this reason, analysis of the "missing disinflation" of recent years has not converged on the role of a flattening in the Phillips curve.
For example, Christiano, Eichenbaum, and Trabandt use a DSGE model very similar to that in Boivin, Kiley, and Mishkin and Del Negro, Giannoni, and Schorfheidebut attribute the modest decline in inflation relative to pre norms to a decline in technological progress, not a flat Phillips curve.
Explaining the inflation surprise via shadow slack A number of researchers have linked the modest disinflation since to the notion that slack is not well-proxied over this period by the unemployment rate.
A fairly large number of researchers suggested that the very large and persistent increase in long-term unemployment, and the possibility that such unemployed persons are detached from the labor market to a sufficient degree that they do not represent slack, may have been a factor that limited the disinflation over the period of interest.
However, the econometric evidence using aggregate U. Total, Short-term, and Long-term Unemployment Note: The long-term unemployment rate is defined as those unemployed for more than 26 weeks divided by the civilian noninstitutional population over age The short-term unemployment rate is the total unemployment rate minus the long-term unemployment rate.
What happens if all these costs are not zero? Now we would find that, there is always some unemployment in a changing economy. It takes time to find the right job at the right wage; changing patterns of demand for goods in a changing technology require changing demands for particular skills, and it takes time to learn new skills. It is also costly to move and, therefore, sometimes worthwhile to wait in the expectation that mobility may not be necessary.
All the above imply that the concept of full-employment is a very ambiguous one, and not a very sensible one, if taken literally to mean that no one is unemployed. Assume we have flexible money wages. However, this does not mean that there is no unemployment. Let us assume that, for a given structure of the labour market, unemployment is uf as in Fig. Because there is excess demand for labour, two things may happen.
First, that the level of unemployment at this real wage is less than uf. Second, that because of the excess demand the money wage will rise. With an excess demand for labour a wider range of skills would be demanded and the employer would be willing to substitute available skills for unavailable ones.
It is also likely that capital would move to areas where labour is available, thus reducing the cost of labour mobility, and information may also spread more quickly. We shall thus assume that there is an indirect relationship between the excess demand for labour and unemployment. Given the excess demand for labour, money wages will rise, until the equilibrium wage rate is reached.
Let us also assume that the rate at which money wages rise depends on the excess demand for labour; the greater the pressure in the labour market the faster is the rate of change of money wages.
We thus have two relationships: Combining these two we get a relationship between unemployment and inflation. This relationship is shown in Fig. The full-employment is represented by uf, where there is no tendency for money wages to change, even though there is some unemployment which is known as natural rate of unemployment. Any level of unemployment less than uf implies that money wages will rise, because there is an excess demand for labour, and the rate at which they will rise depends on the excess demand for labour.
At any unemployment greater than uf there is an excess supply of labour and money wages are likely to fall, at a rate depending on the size of the excess supply. The relationship shown in Fig. Phillips who first discovered the empirical relationship between the change in wages and employment in the U.
So far we have been discussing the relationship between the change in wages and unemployment. To relate the above discussion to an analysis of inflation we have to postulate some relationship between a change in money wages and a change in prices.
Inflation and Unemployment (With Diagram)
There have been various ways in which this has been done, such as, mark up theories of pricing or marginal productivity theory of wage determination.
For our purpose it does not matter what is the exact relationship between the change in money wages and a change in prices. Let us assume that there is a positive relationship between the two. We can then translate Fig. Assume that the government wishes to maintain unemployment at u, less than uf.
This is a more complex development of the first theme discussed above in terms of the comparative static model but it does not incorporate the second theme—expectations and adjustment to them. Expectations and the Phillips Curve: To discuss expectations in the analysis of the Phillips Curve, and to sketch briefly some of the recent developments in this field we start with the labour market again.
The real wage depends on two factors, the money wage and the price level. When there is inflation both the money wage and the price level are changing and both are expected to change to some extent. An individual selling his labour receives a money wage offer and he has to assess what is the real wage represented by this offer.
To do this he has to think what the price level could be over the period for which he offered the money wage. The actual real wage is W0 divided by whatever the price level is expected. We thus have a possibility that the actual real wage may be greater or smaller than the real wage expected by the suppliers of labour on which they base their decision about how much labour to supply.
We shall be interested in the rate of change of the variables and thus. We is the expected change of percentage in real wages. Equation 1 states that, the expected percentage change in real wages is equal to the expected percentage change in money wages minus the expected change in the price level. Now, assume that the expected change in money wages is equal to the actual change-in money wages which is equal to the rate of inflation—the percentage change in prices.
We are also assuming that there is an ongoing inflation in which prices and money wages change at the same rate. Equation 2 says that the expected change in the real wage depends on the difference between the actual rate of inflation and the expected rate of inflation.
If the actual rate of inflation is greater than the expected rate, the expected real wage is increasing; if the actual inflation is equal to expected inflation, the expected real wage remains unchanged; and if the actual rate of inflation is less than the expected rate, the expected real wage is falling. The supply curve shows the relationship between the quantity of labour supplied and the real wage.
However, what is now important for the supply of labour is not the current real wage, but the expected real wage. Only if the expected change in the real wage is zero, is the current real wage the correct variable for the supply of labour. This is shown by the middle supply curve of labour where the expected rate of inflation is equal to the actual rate of inflation.
Again, from equation 2this is the supply curve for which the expected rate of inflation is greater than the actual rate. Consider now that an initial equilibrium in which the actual inflation rate is equal to the expected inflation rate and assume that both are zero no inflation.
The equilibrium in the labour market is at A where real wage is W0 and employment is L0. Associated with this full-employment there is some level of unemployment uf, which is called the natural rate of unemployment.
Assume that, now prices start rising, but at this level expectations about changes in the price level remain unchanged. Thus, the expected change in the price level is less than the actual change in the price level. The supply curve of labour moves to the right where real wage falls but the quantity of labour supplied rises to L1 and unemployment falls. The greater is the rise in prices, the larger is the difference between the actual change in prices and the expected change in prices.
The shift of the supply curve to the right is greater, employment is larger and unemployment is smaller. Let us translate this into a Phillips curve. When the rate of change in prices is zero—and everyone expects it to continue to be zero—equilibrium in the labour market in Fig. L0 with unemployment at uf in Fig. Labour market equilibrium in Fig. Thus, we get the downward sloping Phillips curve. However, it is clear that this Phillips curve was derived on the assumption that the expected change in prices is zero.
Now assume that, people begin to expect that the inflation will continue at some positive rate and the difference between the actual rate and expected rate of inflation will decrease and the supply curve of labour in Fig 16 6 will start shifting back.
Employment will start falling and unemployment rising In Fig. Each of the Phillips curves in Fig. When that expected rate of inflation is equal to the actual rate of inflation, unemployment will be uf. Thus, the vertical Phillips curve at uf shows the relationship between inflation and unemployment when the expected rate of inflation is equal to the actual rate.
Along this curve there is no relationship between the two, and unemployment cannot be changed by increasing the rate of inflation, which is known as the long-run Phillips curve.
There was a problem providing the content you requested
The Phillips curve analysis which takes into account expectations gives some insight not only into the effect of inflation but also into the effects of reducing inflation.
Only after expectations become adjusted to the lower inflation rate docs the Phillips curve starts shifting to the left, and unemployment starts falling. When full adjustment occurs, unemployment will be back at uf and we will be on the dashed Phillips curve. Thus, once expectations about inflation have been built into the system it may be quite costly to reduce the rate of inflation. Inflation and Interest Rates: Now we discuss the link between inflation and interest rates. Economists call the interest rate that the bank pays for our deposits the nominal interest rate and the increase in our purchasing power the real interest rate.
The real interest rate r is the difference between the nominal interest rate i and the rate of inflation. We can show that the nominal interest rate is the sum of the real interest rate and the inflation rate: This is called the Fisher Equation which shows that the nominal interest rate can change for two reasons: The quantity theory shows that the rate of money growth determines the rate of inflation.
The Fisher Equation tells us to add the real interest rate and the inflation rate to determine the nominal interest rate. Thus, the quantity theory and the Fisher Equation together tell us how money growth affects the nominal interest rate.
Two Real Interest Rates: Ex Ante and Ex Post: The real interest rate the borrower and lender expect when loan is made, is called the ex ante real interest rate, and the real interest rate actually realised, is called the ex post real interest rate.