next 12

The average unemployment rate in Japan from 1970 to 2007 was 2.8 per cent, compared with 6.5 per cent in Australia and 6.1 per cent in the United States. If we take the average unemployment rate to be a rough estimate of the underlying natural rate of unemployment, it would appear that, over the period, the natural rate in Japan was not much more than one-half of the Australian and US natural rates. Why the difference?
One of the main reasons appears to have been the widespread reliance on lifetime employment in the Japanese labour market.The typical pattern of working life was one in which new workers quickly settled on a job and kept it until retirement.To give workers incentives to stay in their jobs.Japanese firms offered wages that increased steeply with seniority, relying mostly on seniority-based promotions and offering large lump¬sum retirement payments. In exchange for job security. Japanese workers allowed firms to reassign them to other divisions or even to affiliated companies. When car sales at Nissan declined in the 1980s, for example. Nissan sent some of its workers from idle production lines to the dealerships, to help promote sales.
One implication of these labour arrangements is that flows of workers through the labour market are much smaller in Japan than they are in Australia and the United States. In Japan, there have been no temporary layoffs and many fewer permanent layoffs. A much larger proportion of the needed reallocation of workers took place within firms, rather than through the labour market.
To see why lower flows lead to a lower natural rate of unemployment, think of two countries that are identical in all respects except for the size of the labour-market flows.
In country I (think Japan), 2 per cent of the workers become unemployed every quarter. In country 2 (think Australia). 6 per cent of the workers become unemployed every quarter. In both countries the average duration of unemployment—the average time it takes an unemployed worker to find employment—is the same, one quarter.
Under these assumptions, the unemployment rate in country I (Japan) will be equal to 2 per cent (the flow into unemployment—2 per cent per quarter times a duration of unemployment of one quarter). The unemployment rate in country 2 (Australia) will be equal to 6 per cent (6 per cent per quarter times a duration of unemployment of one quarter).The country with the lower flows will have a lower natural rate of unemployment.
' BOX
Should we expect the natural rate of unemployment to remain very low in Japan? Some economists believe the answer is 'no'. They point out that Japanese firms, facing increasing international competition, are finding it too costly to continue to offer lifetime employment. If this is the case, flows of workers and, by implication, the natural rate of unemployment, are likely to be higher in Japan in the future than in the past. Since 1995. the Japanese unemployment rate rose from 3 per cent to 5.5 per cent at the beginning of 2003, but then fell to 3.9 per cent in 2007. If the trend persists, the natural rate of unemployment in Japan will certainly rise. 
Explaining European unemployment
Thus lar we have ignored Hurope. But il our purpose is to show that the natural unemployment rate can change over time, the evidence from Europe is actually much stronger than lor Australia, the United States and lapan. Recall our discussion ol the evolution ol European unemployment in Chapter I. The European unemployment rate, which until the early 1970s had been much lower than the US rate and similar to the Australian rate has steadily increased. After exceeding 10 per cent in the 1980s and 1990s, it started decreasing, reaching 8 percent in 2001. But in mid-2008 it was 6.8 per cent, compared with -I per cent lor Australia and 5.5 per cent lor the United States.
A high unemployment rate doesn't necessarily reflect a high natural rate of unemployment,- it can reflect instead a large deviation of the actual unemployment rate from the natural rate of unemploy¬ment. How can we tell? Equation > 8.10 ) gives us a clue: by looking at the change in inllation, 77,- 77, II inllation is decreasing last this is an indication that the actual unemployment rate, it,, is far above the natural rate of unemployment ;/„. II inflation is stable, this is an indication that the actual unemploy¬ment rate and the natural rate ol unemployment are roughly equal, and that the natural rate itsell is high, somewhere between 8 percent and 9 percent.
As you saw in Table 1.3 in Chapter I EU countries have roughly stable inflation today, from 1996 to 2008, inflation varied in a narrow range between I per cent and 3.8 per cent per year. This suggests that the actual and the natural rates ot unemployment are roughly equal. Equivalently, the high rate of unemployment in Europe today reflects a high natural rate ol unemployment. This in turn tells us where we should look lor explanations: in the factors that determine the wage-setting and price-setting relations.
Is it easy lo identify the relevant factors? Thc comment is often heard that one ol the main problems ot Europe is its labour-market rigidities These rigidities, the argument goes are responsible lor its high unemployment. While there is some truth to this statement, the reality is more complex.
What do critics have in mind when they talk about the 'labour-market rigidities' afflicting Europe? They have in mind in particular:
• A generous system of unemployment insurance. Thc replacement rate—that is, lhe ratio ol unemploy¬ment benefits to the alter-tax wage—is often high in Europe, and the duration ol benefits—the period ol time lor which the unemployed are entitled ю receive benefits often runs in years. Some unemployment insurance is clearly desirable. But generous benclits may increase unemployment in at least iwo ways. They decrease the incentives the unemployed have it) search lor jobs. They may also increase the wage that lirms have to pay. Recall our discussion ol efficiency wages in Chapter 6: the higher unemployment benefits arc, the higher the wages lirms have to pay in order to motivate and keep workers.
• A high degree of employment protection. By employment protection, economists have in mind lhe set ol rules that increase the cost ot layoffs tor firms. These range from high severance payments, lo the need lor firms to justify layoffs, to thc possibility tor workers to appeal the decision and have ii reversed. I he purpose ol employment protection is to decrease layolfs and thus to protect workers Irom the risk ol unemployment. What it does however is also increase the cost ol labour lor firms, thus reducing hires and making it harder lor the unemployed to get jobs. The evidence suggests that while employment protection does not necessarily increase unemployment, it changes its nature. The llows in and out ol unemployment decrease, but the average duration of unemployment increases. Such a long duration increases the risk that the unemployed lose skills and morale, decreasing their employability.
• Minimum wages. Most European countries have national minimum wages. In some countries the ratio ol the minimum wage to the median wage can be quite high High minimum wages clearly run the risk ol decreasing employment for the least skilled workers, thus increasing their unemployment rate.
4 Look at the change in inflation to infer whether high unemployment reflects a high natural rate of unemployment or unemployment above the natural rate of unemployment. From equation (8.10): If 77, - 77,< 0. it must be that u, > un If 77, - = 0, it must be that u, = u„
THL NATURAL WE ОГ UNEMPLOYMENT AND THE PHILLIPS CURVE
chapter 8
• Bargaining rules In most European countries, labour contracts are subject lo extension agreements. A contract agreed to by a subset ol firms and unions can be automatically extended lo all lirms in the sector. This considerably reinlorces the bargaining power ol unions because it reduces the scope
for competition bv non-unionised firms. As we saw in Chapter 6, stronger bargaining power on ihe part of the unions may result in higher unemployment. Higher unemployment is needed to reconcile the demands of workers with the wages paid by firms.
Do these labour-market institutions really explain high unemployment in Europe? Is the case open
and shut? Not quite. Here it is important to recall two important lacts.
• Fuel 1: As we saw in Chapter I unemployment was not always high in Hurope. Recall the evolution ot unemployment shown in Table 1.3. In the 1960s, the unemployment rate in the four major continental European countries was lower than that in the United States, around 2-Я per cent. The natural rate in these countries today is around 8-9 per cent. How do we explain this increase? One hypothesis is thai institutions were dillerent then and that labour-market rigidities have appeared only in the past forty years. This turns out not to be the case, however It is true that, in response to the adverse shocks of the 1970s in particular the two recessions lollowing the increases in the price ot oil), many Huropean governments increased the generosity ol unemployment insurance and the degree of employment protection. But even in the 1960s, European labour-market institutions looked nothing like LIS labour-market institutions. Social protection was higher in I urope, yet unemployment was lower.
A different line of explanation focuses on the interaction between institutions and shocks. Some labour-market institutions may be benign in some environments and very costly in others. Take employment protection. If competition between lirms is limited, the need to adjust employment in each firm may be limited as well and so the cost ol employment protection mav be low. Hut il competition, Irom either domestic firms or foreign firms, increases the cost of employment protection may become very high, l irms that cannot adjtist their labour lorce quickly mav simply be unable to compete and may go out of business. Thus, even if employment protection rules do not change, higher competition can lead to a higher natural rate.
• Fact 2: Many European countries actually have a low unemployment rate. This is shown in I igure K.6 which gives the unemployment rate tor fifteen Huropean countries the 15 members ol the European Union before the increase in membership to 27'. In all these countries inflation is stable so the unemployment rate is roughly equal to the natural rate. The unemployment rate is high in the four large continental countries i which is why we focused on them in Chapter I But note how low the unemployment rate is in some other countries—in particular, Denmark Ireland and the Netherlands

 
Is it the case that these low-unemployment countries have low benefits, low employment protection and weak unions? Things are unfortunately not so simple. Countries such as Ireland and the United Kingdom do have labour market institutions that resemble those of the United States: limited benelits low employment protection and weak unions. But countries such as Denmark and the Netherlands have a high degree of social protection —in particular, high unemployment benefits and strong unions.
More concretely, when inflation runs on average at 5 per cent a year, wage setters can be ^ confident that inflation will be between 3 per cent and 7 per cent. When inflation runs on average at 30 per cent a year, wage setters can be confident that inflation will be between 20 per cent and 40 per cent In the first case, if they set a nominal wage their real wage may end up 2 per cent higher or lower than they expected: in the second case, it may end up 10 per cent higher or lower than they expected. There is much more uncertainty in the second case.
This assumption is too strong. Indexation clauses typically adjust wages not for current inflation (which is only known with a lag), but for inflation in the recent past, so the-e remains a short delay between inflation and wage adjustments.We ignore this delay here.
So what is i >ne to conclude? An emerging consensus among economists is that the devil is in the details. Generous sot ial protection is consistent with low unemployment but it has to be provided efficiently. For example unemployment benefits can be generous as long as the unemployed are at the same time, forced to take such jobs as are available. Some employment protection—(or example, in the form of generous severance payments—may not prevent low unemployment as long as firms do not lace the prospect oi long administrative or judicial uncertainty when they lay off workers. Countries such as Denmark appear to have been more successful in achieving these goals. Creating incentives for the unemployed to take jobs and simplifying the mles ol employment protection are on the reform agenda ol many European governments, in the hope that this will lead to a decrease in the natural rate in the future.
High inflation and the Phillips curve relation
Retail how in the 1970s, the Phillips curve changed as inflation became more persistent and wage setters changed the way they formed expectations. The lesson is a general one: the relation between unemployment and inflation is likely to change with the level and the persistence of inflation. Evidence from countries with high inflation confirms this lesson. Not only does the way workers and lirms form their expectations change, but so do institutional arrangements.
When the inflation rale becomes high, inflation also tends to become more variable. As a result, workers and firms become more reluctant to enter into labour contracts that predetermine nominal wages lor a long period of time. If inflation turns out to he higher than expected, real wages may plunge and workers may suffer a large cut in their standard ol living. It inflation turns out to be lower than expected, real wages may go up sharply, l irms may not be able to pay their workers, and some lirms may become bankrupt.
lor this reason, the form of wage agreements changes with the level of inflation. Nominal wages are set lor shorter periods ot time, down from a year to a month or even less. Wage indexation, a rule that automatically increases wages in line with inflation, becomes more prevalent.
Wage indexation has been part ol Australia's relatively recent history. We saw in Chapter I that inflation rose quite dramatically in Australia in the mid-1970s, due to the large ОРГС oil price hike and because the institutionalised award wage system in Australia delivered a very large wage increase in 1974. Natural s' inflation 'and unemplovmentl rose, reaching a post-Korean War peak ol nearly 20 per cent. In response the Industrial Relations Commission introduced wage indexation in March 1975, which remained in place lor four years.
I he introduction ol wage indexation leads, in turn, to a stronger response of inflation to unemploy¬ment To see this an example based on wage indexation will help. Think of an economy that has two types of labour contracts, A proportion A 1 the Creek lowercase lambda >, of labour contracts is indexed: nominal wages in those contracts move one for one with variations in the actual price level. A propor¬tion 1 A ol labour contracts isn't indexed: nominal wages are set on the basis ol expected inllation. Expected inflation is equal to last years inllation.
Under this assumption, equation 8.91 becomes
77, = [ А 7Г, + (I - A)irf] - tt(M, - !/„)
THE NATURAL RATE Ol UNEMPLOYMENT AND THE PHILLIPS CURVr
chapter 8
I he term in brackets on the right reflects the tact that a proportion (A' of contracts is indexed and thus responds to actual inflation • and a proportion I - A responds to expected inflation, 77'',. It we
77, ,), we get
assume that this year's expected inflation is equal to last year's actual inflation 177''. = , >
77, = [ A 77, + (I - A) 77,. | | - a(U, - u„)
(8.И)
When A = 0, all wages arc set on thc basis ol expected inllation (which is equal to last year's inflation, 7rf_| i and the equation reduces to equation (8.10):
7Г, - щ | = - a(u, - u„)
When A is positive, however, a proportion A ol wages is set on the basis ol actual inllation rather than expected inflation. To see what this implies, reorganise equation (8. I I move thc term in brackets lo the left, factor (I - A1 on lhe left ol lhe equation, and divide both sides by I A to get
7Г, - 7Г. , = ^ (M, - u„)
(1-A)
Wage indexation increases the effect of unemployment on inflation. The higher the proportion ol wage contracts that arc indexed—the higher A—the larger the effect ol the unemployment rate on the change in inllation—the higher the coefficient a/{ I - A .
Thc intuition is as follows. Without wage indexation lower unemployment increases wages which in turn increases prices. But because wages dont respond to prices right away, there is no further ellect within ihe year. With wage indexation, however, an increase in prices leads to a further increase in wages within the year, which leads to a further increase in prices, and so on so that ihe ellect ol unemployment on inllation within the year is higher,
II and when A gets close lo I—which is when most labour contracts allow lor wage indexation— small changes in unemployment can lead to very large changes in inllation. Put another way, there can be large changes in inflation with nearly no change in unemployment. This is what happens in countries High inflation is the where inllation is very high: the relation between inllation and unemployment becomes more and more topic of Chapter 24. ► tenuous and eventually disappears altogether.


Deflation and the Phillips curve relation
We have just looked at what happens to the Phillips curve when inflation is very high. Another issue is what happens when inflation is low, and possibly negative- when there is deflation.
The motivation for asking the question is given by an aspect ol Figure 8.1 we mentioned at the start ol the chapter but then left aside. In thai ligure note how the points corresponding to the 1930s (they are denoted by triangles) lie to the right ol thc others. Not only is unemployment unusually high- -this is no surprise as we are looking at thc years corresponding to the Crcat Depression—but given the high unemployment rale, the inflation rale is surprisingly high In other words, given ihe very high unemploy¬ment rate, we would have expected not merely deflation but a large rate ol deflation. In tact deflation was limited, and from 19.33 to 1935 inflation was actually positive.
How do we interpret that fact? There are two potential explanations.
One is that the Great Depression was associated with an increase not only in the actual unemploy¬ment rate but also in the natural unemployment rale. This seems unlikely. Most economic historians see¬the Depression primarily as the result ol a large adverse shilt in aggregate demand thus as an increase- in the actual unemployment rate over thc natural rate ol unemployment, rather than an increase in the natural rale ol unemployment.
Thc other is that when the economy starts experiencing deflation, the Phillips curve relation breaks down. One possible reason is the reluctance ol workers to accept nominal wage cuts. Workers may be willing lo accept a cut in real wages thai comes Irom nominal wages increasing more slowly than inflation they may, however, light the same cm in real wages il il comes with an absolute decrease in nominal wages 11 this argument is correct, this implies thai the Phillips curve relation between the change in inllation and unemployment may disappear, or at least become weaker, when the economy is close to zero inflation.
Consider two scenarios: In one. inflation is 4 per cent and your nominal wage goes up by 2 per cenc. In the other, inflation is 0 per cent and your nominal wage is cut by 2 per cent. Which do you dislike the most? You should be indifferent between the two. In both cases, your re a wage goes down by 2 per cent.There is some evidence, however, that most people find the first scenario less painful. More on this in Chapter 26. ►
For more on the Japanese economic slump, see Chapter 23. ►
This issue is a crucial one at this time because in many countries inflation is now verv low and tailing. As you saw in Chapter I. Japan has actually been having negative inllation. What happens lo the Phillips curve relation in this environment ol low inflation or even deflation is one ot the developments closely watched by macroeconomists today.
THE NATURAL RATE Ob UNEMPLOYMENT AND - HE PHIll IPS CURVE
chapter 8
SUMMARY
• The aggregate supply relation can he expressed as a relation between inflation, expected inflation and unemployment. The higher the expected inflation, the higher is actual inflation. The higher the unemployment, the lower is inllation.
• When inllation isn't very persistent, expected inllation doesn't depend very much on past inflation. Thus, the aggregate supply relation becomes a relation between inflation and unemployment. This is what Phillips, lor the United Kingdom and Australia, and Solow and Sanntelson, for the United States, discovered when they looked in the late 1950s, at the joint behaviour ol unemployment and inflation.
• As inflation became more persistent in the 1970s and 1980s, expectations ol inflation became based more and more on past inflation. The modern aggregate supply relation lakes the form of a relation between unemployment and the change in inllation. High unemployment leads to decreasing inflation; low unemployment leads to increasing inflation. Other factors, such as foreign inllation, may alfect the dynamics ot inflation in an open economy.
• The natural unemployment rate is the unemployment rate at which the inflation rate remains constant. When the actual unemployment rate is above the natural rate ol unemployment, the inllation rate decreases,- when the actual unemployment rate is below the natural unemployment rale, the inflation rale increases.
• The natural rate ol unemployment depends on many factors that differ across countries and can change over lime. This is why the natural rale ol unemployment varies across countries. It is much lower in Japan than in Australia or the United States. Also, the natural unemployment rate varies over time. In Australia, the natural unemployment rale probably increased Irom about 2 per cent in the 1960s and the early 1970s lo well above 6 per cent through the early 1990s, and was probably close to 5.5 per cent in 2008. In P.urope, the natural unemployment rate has increased a lot since the 1960s—by about 4 per cent. In Japan, ii has increased significantly in the lasi twenty years.
• Changes in the way the inflation rate varies over time allect the way wage setters form expectations and also how much they use wage indexation. When wage indexation is widespread, small changes in unemployment can lead to very large changes in inflation. Ai high rates of inflation, the relation between inflation and unemployment disappears altogether.
• At very low or negative rates ol inflation, the Phillips curve relation appears to become weaker. During the Great Depression, even very high unemployment led only to limited deflation. The issue is important because many countries have low and falling inflation today.