OKUN'S LAW ACROSS COUNTRIES AND TIME

Let s write equation (9.2 1 using letters rather than numbers. Lei g, denote the normal growth rate ol the economy i about 3.2 per cent per year lor Australian Let the coefficient (3 ihc (.reek lowercase beta measure the effect of output growth above normal on the change in the unemployment rate. iAs you saw in equation (9.2 in Australia /3 equals 0.5. The evidence lor other countries is given in the locus box 'Okun's law across countries and time ) We can then write:
Щ - Им = - Йр
Output growth above normal leads to a decrease in the unemployment rate,- output growth below normal leads to an increase in the unemployment rate.


The coefficient fi in Okun's law gives the effect on the unemployment rate of deviations of output growth from normal. A value for (3 of 0.5 tells us that output growth I per cent above the normal growth rate for one year decreases the unemployment rate by 0.5 per cent.
The coefficient /У depends in part on how firms adjust employment in response to fluctuations in their production. This adjustment of employment depends in turn on such factors as the internal organisation of firms and the legal and social constraints on hiring and firing. We would expect the coefficient to be different across countries, and it is. Table I gives the estimated coefficient fi for a number of countries.
The first column gives estimates of /3 based on data from 1960 to l980.The United States has the highest coefficient. 0.39, followed by Australia, 0.24. Germany. 0.20, the United Kingdom, 0.15 and Japan, 0.02.
The ranking in the first column fits well with what we know about the behaviour of firms and the structure of firing and hiring regulations across countries. /3 is smallest in Japan. As you saw in Chapter 8, Japanese firms offer a high degree of job security to their workers, so variations in output have little effect on employment and so also little effect on unemployment, fi is largest in the United States, where there are few social and legal constraints on firms' adjustment of employment. And legal restrictions on firing—from severance pay to the need for legal permission from the state to terminate employment—explain why the coefficients estimated for the two European countries are in between those of Japan and the United States. Australia was only marginally more flexible than the European countries.

OKUN'S LAW ACROSS COUNTRIES AND TIME
The second column gives estimates based on data from 1981 to 2006. The coefficient is changed only slightly for the United States, but is higher for the other four countries.This again fits with what we know about firms and regulations. Increased competition in goods markets since the early 1980s has led firms in most countries to reconsider and reduce their commitment to job security. And. at the urging of firms, legal restrictions on hiring and firing have been considerably weakened in many countries. Both factors have led to a larger response of employment to fluctuations in output, thus to a larger value of /3. Interestingly. Australia and the United Kingdom have become the most flexible in this context.
Table 1 Okun's law coefficients across countries and time
1
Country 1960-80 1981-2006
1
Australia 0.24 0.49
United States 0.39 0.42
United Kingdom 0.15 0.51
Germany 0.20 0.29
Japan 0.02 0.1 1
SOURCE Authors' calculations.
 
The Phillips curve: From unemployment to inflation change
(9.41
We derived in Chapter 8 the lollpwing relation between inllation, expected inflation and unemploy¬ment (equation (8.7! :
«(», - H„I
Inflation depends on expected inflation and on the deviation ol unemployment from the natural rate of unemployment.
We then argued that, in Australia today, expected inflation is well approximated by last year's inllation, so that we can replace тт1', by 7T,_,. With this assumption, the relation between inflation and unemployment takes the form


(9.5)
H„)
- 7Ti_


Phillips curve: ► u, < u„ 7Г, > 7rHl u,>u„=> 7r,< jrH
Unemployment below the natural rate leads to an increase in inllation,- unemployment above the natural rate leads to a decrease in inflation. The parameter cr gives the ellect ol unemployment on the change in inflation. You saw in Chapter 8 that, since I97U in Australia, thc natural unemployment rate has been on average equal to 6.25 per cent and a equal to 0.32. As we explained earlier, these numbers do change a little depending on the period you use. In this chapter we are focusing on thc period from 1982 to 2007, and for that period we gel the following relation:


(9.6)
TT, - 7T._|
-0.26(«, - 6.5%)


The value ol a equal to 0.26 means lhat an unemployment raie ol I per cent above lhe natural rate lor one year leads to a decrease in the inflation rate ol 0.26 per cent, while the natural rate is estimated as 6.5 per cent.
The dynamic aggregate demand relation: From changes in inflation to output growth
We derived in Chapter 7 the aggregate demand relation between output and the price level based on equilibrium in goods and financial markets—equation (7.3) when the central bank kept the money stock fixed and equation (7.5) when the central bank used an interest rate rule. These AD relations applied to a situation when the price level was constant and output reached the natural rate in the medium run. This is unrealistic tor two reasons:
• Inflation is tvpically not zero in the medium mn. and, in thc case where thc central bank has an interest rale rule, it is usually based on achieving a non-zero inflation target.
• Output typically grows at what we have already called the normal rate x„ in the subsection above- on Okun's law.
Whet we want to do now is set up a dynamic aggregate demand curve, which relates the inflation rate and output growth-—in both the short run and the medium run. The basic idea is simple: it inflation rises, the market or the central bank responds with a rise in the interest rate, which will reduce the rare of growth ol aggregate demand and thus output. In the medium am, there will he positive output growth and inflation, but no trade-off between them, fortunately, it is easy to derive such a relation by making a few convenient assumptions. First, we derive a dynamic version ol the IS relation, then we derive a dynamic version ol monetary policy rule, and then we put them together to give a dynamic aggregate demand relation.
Chapter 26 provides ► details on the design of monetary policies around the world and over time. Have a quick look now at the section that discusses inflation targeting.
Lets begin with the case where the central bank keeps control ol nominal money. Then we will study the more relevant case lor today in Australia, where the central bank has an interest rate rale. Though we will focus on the latter in the subsequent analysis, you will see lhat the dynamic aggregate demand relations derived lor each are quite similar.