In the early 1930s, more than one quarter of the UK labour force was unemployed. In particular regions and occupations the unemployment rate was very much higher. [employment of economics teachers was hardly effected at all]. High unemployment means that the economy is throwing away output by failing to put its people to work. It also means misery, social unrest, and hopelessness for the unemployed. Over the following 40 years, macroeconomic policy was geared to avoiding a rerun of the 1930s. Figure 26-1 shows that it succeeded.
In the 1970s views about unemployment bagan to change. People began to reject the Keynsian pessimism about the capacity of the economy to respond to shocks by quickly restoring full employment. The classical model began to be more widely accepted as a description of the way the economy works even in the relatively short run. Since in the classical model unemployment is voluntary there is less presumption that unemployment means extreme human suffering. Moreover, research by labour economists has shown that in the 1950s and 1960s most of the unemployed quickly found jobs. Unemployment might therefore be considered a stepping stone to a better job.
By the 1970s, not only was it felt that the cost of unemployment might have been overstated; in addition, governments in many countries began to percieve an even greater danger to economic and social stability, the danger of high and rising inflation. Thus by the end of the 1970s many governments had embarked on tight monetary and fiscal policies into try to keep inflation under control. The combination of restrictive demand policies and the adverse supply shock of hte second major OPEC oil price increase in 1979-80 has led to a dramatic increase in unemployment in most of the industrial countries in the early 1980s. Figure 26-1 shows data for the UK. Data for other countries are shown later in the chapter.
High unemployment is one of the major problems of the 1980s. Will it contine? Is it a drain on society or a signal that iat lst people are getting out of dead end jobs into something better? What can and should the government be doing? These are the questions we set out to answer in this chapter. We begin by looking at the facts.
26-1 THE FACTS
Not everyone wants a job. Those who do are called the labour force.The participation rate is the percentage of the population of working age who declare themselves to be in the labour force. In Chapter 10 we pointed out that postwar growth of the UK labour force had been caused by an increase in the population of working age who declare themselves to be in the labour force. In Chapter10 we pointed out that postwar growth iof the UK labour force has been caused less by an increase in the population of working age than by an increase on paricipation rates, most noticably by married women.
The unemployment rate is the percentage of the labour foce who are withoug a job but are registered as being willing and vailable for work.Of course some people without a job are really looking for work but have not bothered to register as unemployed. These people will not be included in the official statistics for the registered labour forcem nor will they appear as registered unemployed. Yet from an economic viewpoint, such people are unemployed. This is an important phenominon to which we return shortly. For the momentm when we present evidence on the size of the labour force or the number of people unemployedm it should be undestood that data refer to the registered labour force and the registered unemployed.
Figure 26-1 makes two main points about the unemployment rate in the UK.
First, unemployment was high during the interwar years, especially during the great recession of the 1930s. It was the persistance of unemployment that led Keynes to develop his General Theory.
Second, by comparison, post war unemployment was tiny until the late 1970s. By the early 1980s it was starting to get back to prewar levels. This basic pattern applies in many other industrialised countries. [I doubt this is true - JR]
Stocks and Flows
Unemployment is a stock concept measured at a point in time. Like a pool of water, its level rises when inflows (the newly unemployed) exceed outflows (people getting new jobs or quitting t he labour force altogether). Figure 26-2 illustrates this important idea. Beginning with people working, there arc three ways to become unemployed. Some people are sacked or made redundant (job-losers); some are temporarily laid off but expect eventually to be rehired by the same company; and sonic people voluntarily quit their existing jobs. But the inflow to unemployment can also come from people not previously in the labour force: school-leavers (new entrants), and people who once had a job, then ceased even to register as unemployed, and are now coming back into the labour force in search of a job (re-entrants). People leave the unemployment pool in the opposite directions. Some get jobs. Others give up looking for jobs and leave the labour force completely. Although some of this latter group may simply have reached the retirement age at which they can draw a pension, many of them are discouraged workers, people who have become depressed about the prospects of ever finding a job and decide to stop even trying. Between January 1980 and January 1983, the number of people registered as unemployed in Britain rose from 1.3 million to 3.1 million. Data collected through Department of Employment
jobccntres account for most of this increase and .11 shown in Table 26-1. The table makes the 1)( wit that the pool of unemployment is not skignant. Even with 3 million unemployed, this tit her is less than the number of people entering and leaving the pool every year.
I I )nr.itioll thiciiiplovnicnt When un-employment is high, people have to spend longer in the pool before they find a way out.
Table 26-2 gives data on the duration of unemployment. As the level of unemployment has risen, the problem of the long-term unemployed has returned. Whereas in 1974 only 20 per cent of the people unemployed had been out of work for longer than one year, by April 1983 this pro-portion had risen to 36 per cent. And over the same period, the fraction of the unemployed who had been unemployed for less than eight weeks fell from 44 to 17 per cent. Unemployment can no longer be regarded as a temporary stopover on the way to better things. 
26-2 THE FRAMEWORK
Having introduced some of the most important facts about unemployment in the UK, we now develop a theoretical framework in which to discuss the subject. We begin with the old-style classification of types of unemployment, which emphasizes the source of the problem. Then we discuss the modern approach to unemployment which emphasizes the way people in the labour market are behaving.
Types of Unemployment Economists used to classify unemployment a frictional, structural, demand-deficient, or clamp ical. We discuss each in turn.
Frictional Unemployment
This is the irredrit ible minimum level of unemployment in .1 dynamic society. It includes people wIr.)4e physical or mental handicaps make them aim, 'NI unemployable, but it also includes the pc(11►1, spending short spells in unemployment Ir. ■ hop between jobs in an economy where b( )111 r h. labour force and the jobs on offer are cont in tr.111‘ changing.
Structural Unemployment
In the longer run, the pattern of demand and production is alwir‘ 4 changing. In Chapter 31 we discuss the re.r,...114 why particular countries in the world CC(111411M come to specialize in the production of 1).11 ticular commodities at particular times. In re, (Au decades industries such as textiles and heavy engineering have been declining in the UK, Structural unemployment refers to unemployment arising because there is a mismatch oi and job opportunities when the pAtIci II of
Itint.t i id and production changes. For example, a 44111C(I welder may have worked for 25 years in %Ittplwilding but is made redundant at 50 when the industry contracts in the face of foreign competition. That worker may have to retrain in it new which is more in demand in today's et 'II( )111y. But firms may be reluctant to take on mid it din older workers. Such workers become victims of structural unemployment.
Demand-deficient Unemployment
This refers Keynesian unemployment, when aggregate dentaild falls and wages and prices have not yet 'Owed to restore full employment. Aggregate Jett l id is deficient because it is lower than full-iv merit aggregate demand. In chapter 25 we saw that, until wages and have adjusted to their new long-run equilibrium level, a fall in aggregate demand will lead lower output unemployment. Some workers will want to work at the going real wage rate but will he unable to find jobs. Only in the longer run ill wAr,e,, And prices fall enough to boost the
real money supply and lower interest rates to the extent required to restore aggregate demand to its full-employment level, and only then will demand-deficient unemployment be eliminated.
Classical Unemployment Since the classical model assumes that flexible wages and prices maintain the economy at full employment, classi-cal economists had some difficulty explaining the high unemployment levels of the 1930s. Their diagnosis of the problem was partly that union power was maintaining the wage rate above its equilibrium level and preventing the required adjustment from occurring. Classical unemployment describes the unemployment created when the wage is deliberately maintained above the level at which the labour supply and labour demand schedules intersect. It can be caused either by the exercise of trade union power or by minimum wage legislation which enforces a wage in excess of the equilibrium wage rate. The modern analysis of unemployment takes
the same types of unemployment but classifies them rather differently in order to highlight their behavioural implications and consequences for government policy. Modern analysis stresses the difference between voluntary and involuntary unemployment.
The Natural Rate of Unemployment Figure 26-3 shows the market for labour. The labour demand schedule LD slopes downwards, showing that firms will take on more workers at a lower real wage. The schedule LF shows how
many people want to be in the labour force at each real wage. We assume that an increase in the real wage increases the number of people wishing to work. The schedule AJ shows how many people accept job offers at each real wage. The schedule lies to the left of the LF schedule, both because some people are inevitably between jobs at any instant, and because a particular real wage may tempt some people into the labour force even though they will accept a job offer only if they find an offer with a rather higher real wage than average. Labour market equilibrium occurs at the point E. The employment level N* is the equilibrium or full-employment level. The distance EF is called the natural rate of un-employment. The natural rate of unemployment is the rate of unemployment when the labour market is in equilibrium. This unemployment is entirely voluntary. At the equilibrium real wage w"-, N, people want to be in the labour force but only N"- want to accept job offers; the remainder don't want to work at the equilibrium real wage. Which of our earlier types of unemployment must we include in the natural rate,of unemploy-ment? Certainly all frictional unemployment. But we should also include structural unemploy. ment. Suppose a skilled welder earned £150 a week before being made redundant. The issue is not why the worker became redundant (the decline of the steel industry), but why the worker refuses to take a lower wage as a dishwasher in order to get a job, or why steelworkers as a whole did not take a sufficient wage cut to allow the steel industry to remain profitable and com+ petitive at its former levels of output and em-ployment. If the answer is that steelworkers refuse to accept that the equilibrium wage for their skill has fallen, and refuse to work at wages lower than those to which they have been accus-tomed, then we must count this unemploymen as voluntary and include it in the natural rate. They are not prepared to work at the going wag rate but still want to be considered part of tlic. labour force.
What about classical unemployment, for ex-., mple where unions maintain wages above their equilibrium level? This is shown in Figure 26-3 a wage rate w2 above w"-. Total unemploy-ment is now given by the distance AC. As indi-v iduals, a number of workers AB would like to ta ke jobs at the wage rate w2 but will be unable o find them since firms will wish to be at the point A. As individuals, these workers are in-voluntarily unemployed. A worker is involuntarily unemployed if he or she would accept a job offer at the going wage rate. lowever, through their unions, workers collec-t i v el y decide to opt for the wage rate w2 in excess 1 the equilibrium wage, thereby reducing the level of employment. Hence for workers as a w hole we must regard the extra unemployment .1., voluntary. Thus we also include classical imemployment in the natural rate of unemploy-ment. If in the long run unions maintain the w .1 gc lv,, the economy will remain at A and AC is the natural rate of unemployment. 'Ilk leaves only Keynesian or demand-deficient mei i t ployment. Such unemployment is involun-ta r v, being caused by sluggish labour market ustment beyond the control of individual workers or unions. Thus we can divide total unemployment into the equilibrium or natural ate the equilibrium level determined by normal labour market turnover, structural mi%itiatch, union power, and incentives in the 1,111( 1111. market — and Keynesian unemployment, ►ict lilies called demand-deficient or cyclical unemployment — the disequilibrium level of lily( )1tintary unemployment caused by the com-hin.ition of low aggregate demand and wage oilinstment which is sluggish for the reasons we t.N.iiiiined in the previous chapter. l'his division helps us think clearly about the guvernment policies required to tackle the un-employment problem. Since we have argued that fit the long run the economy will gradually niaiLage to get back to full employment through slow process of wage and price adjustment, Keynesian unemployment will eventually get rid
of itself. But in the short run, Keynesian un-employment is the part of total unemployment that the government could help mop up by using fiscal and monetary policy to boost aggregate demand, rather than waiting for wage and price reductions to increase the real money supply and lower interest rates. In contrast, the natural rate of unemployment tells us the part of unemployment that will not be eliminated merely by restoring aggregate demand to its full-employment level. The natural rate is the 'full-employment' level of unemployment. To reduce the natural rate, supply-side policies operating on labour market incentives will be needed. This is the framework we employ for the rest of the chapter. We begin by investigating the large increase in unemployment over the last decade, in order to understand its causes more fully. Then we discuss the prospects for unem-ployment during the rest of the 1980s and the policy options open to the government.
26-3 WHY IS UNEMPLOYMENT SO HIGH?
By 1983 the UK unemployment rate was more than eight times as high as it was in 1965. The task for empirical economists is to try to say how much of this increase was caused by an increase in the natural rate of unemployment and how much was caused by deficient demand and slugg-ish wage adjustment. In Table 26-4 we give some recent estimates of the forces at work, based on the work by Professor Steve Nickell of the London School of Economics. Nickell's esti-mates were derived from data on male unemployment, which is more comprehensively and reliably documented than the unemployment of women.'' For other attempts to estimate the natural rate of un-employment in the UK, see R. A. Batchelor and T. D. Sheriff, `Unemployment and Unanticipated Inflation in the UK', Economica, 1980; and Patrick Minford, Unemployment: Cause and Cure, Martin Robertson, 1983.
In considering the prospects for unemploy-ment in the rest of the 1980s, we begin by looking at how the government could get the natural rate of unemployment down to a lower level. Then we consider how quickly Keynesian unemployment could be reduced.
26-4 SUPPLY-SIDE ECONOMICS
Keynesians believe that the economy can deviate from full employment for quite a long time, cer-tainly for a period of several years. Monetarists believe that the classical full-employment model is relevant much more quickly. But everyone agrees that in the long run the performance of the economy can be changed only by affecting the level of full employment and the correspond-ing level of potential output. Supply-side economics is the use of micro-economic incentives to alter the level of full employment, the level of potential output, and the natural rate of unemployment. Although in this section we are interested chiefly in how to change the natural rate of unemploy-ment, it is convenient to discuss some of the wider implications of supply-side economics at the same time. We return to the determination of potential output in Chapter 29.Income Tax Cuts One of the key themes of supply-side economists is the benefits. that stem from reducing the marginal rate of income tax. The marginal rate of income tax is the fraction of each extra pound of income that the government takes in income tax. We discussed tax rates and work incentives in detail in Chapter 10. We pointed out that a cut in marginal tax rates, and a consequent increase in the take-home pay derived from the last hour's work, tend to make people substitute work for leisure. But against this substitution effect must be set an income effect. To the extent that people now pay less in taxes, they will have to do less work to obtain any given target living standard.
Thus, theoretical economics cannot prove that income tax cuts increase the desired labour supply, and in fact most empirical studies con-firm that, at best, tax cuts lead to only a small increase in the supply of labour. We gave some details in Chapter 10 and give some more in Box 26-2. Figure 26-5 may be used to analyse the effect of a cut in marginal tax rates. The labour demand schedule LD shows that firms demand more workers at a lower real wage. We draw a steep schedule LF showing that higher after-tax real wage rates, at best, lead to only a small increase in the number of people wishing to be in the labour force. The schedule AJ shows how many
FIGURE 26-5 A CUT IN MARGINAL INCOME TAX RATES.
An income tax makes the net-of-tax wage received by households lower than the gross wage paid by firms. When the vertical distance AB measures the amount each worker pays in income tax, equilibrium employment is N,, the quantity that households wish to supply at the after-tax wage w, and that firms demand at the gross wage w. At the after-tax wage w, the natural rate of unemployment is the horizontal distance BC. If income tax were abolished, equilibrium would be at E. Employment would rise from N, to N, and the natural rate of unemployment would fall from BC to EF. Relative to the fixed level of unemployment benefit, the rise in take-home pay from w, to w, reduces the level of voluntary unemployment.
wi
LD
Ni N2 Number of workers
people wish to accept job offers at each real wage. It is drawn for a given (real) level of unemployment benefit. Hence the horizontal distance between the AJ and LF schedules — the it umber of people in the workforce refusing to work at each real wage, or the amount of voluntary unemployment — decreases as the real wage rises relative to the given level of un-employment benefit. Thus the figure incorporates lie fact that a reduction in the replacement ratio, t he ratio of unemployment benefit to wage rates, reduces voluntary unemployment. Suppose initially that there is a marginal 11 come tax rate equal to the vertical distance AB. lhe equilibrium level of employment will then he N, . Why? Because income tax drives a wedge between the gross-of-tax wages paid by firms the net-of-tax wages received by workers. At t lie employment level N, firms are happy to hire this quantity of labour at the gross wage w,. Subtracting the income tax rate AB, N, workers kN'allt to take job offers at the after-tax wage w3. I 'II us N, is the equilibrium level of employment. l'he horizontal distance BC shows the natural ,it e of unemployment, the number of workers in I In' labour force not wishing to work at the going r.i I e of take-home pay. To show the effect of a cut in marginal tax ales, suppose that income taxes were abolished. gross wage and the take-home pay now oiiicide, and the new labour market equilibrium IL, at L. Note that two things have happened. First the equilibrium level of employment has ri wit. Second, although more people wish to be III the labour force because take-home pay has increased from w, to w2, the natural rate of unemployment has fallen from the distance BC t > the smaller distance EF. A rise in take-home pay relative to unemployment benefit reduces I he level of voluntary unemployment. Si nil lar effects would be obtained if, instead of cti t t i lig income tax, the level of unemployment benefit were cut. For a given labour force schedule :1;, fewer people would now wish to he un-employed at any real wage. I fence the schedule (I I, showing acceptances of job of furs, would
UNEMPLOYMENT 599
shift to the right. Again, the effect would be both to increase the equilibrium level of employment (and hence of potential output) and to reduce the natural rate of unemployment by reducing the replacement ratio. What about the effect of changes in the national insurance contributions paid both by firms and by workers? These are mandatory contributions to state schemes which provide unemployment and health insurance. They act like an income tax in driving a wedge AB between the total cost to a firm of hiring another worker and the net take-home pay of a worker. Figure 26-5 shows that a reduction in these contributions will increase the equilibrium level of employment, increase the equilibrium level of take-home pay, reduce the replacement ratio, and reduce the natural rate of unemployment.
Other Policies Aimed at Labour Supply In Figure 26-3 we showed that, by restricting labour supply, unions could force firms up their labour demand schedule. In consequence, the equilibrium real wage would be higher but the equilibrium level of employment lower. Since a higher real wage reduces employment but (slightly) increases the number of people wishing to be in the labour force, we said that in raising real wages unions had increased the natural rate of unemployment. Collectively, labour had opted for higher wages and more unemployment. Conversely, the natural rate of unemployment will be reduced if the power of organized labour is weakened. Unions will then be less successful in restricting labour supply and forcing up wages. Hence any government intervention in the labour market to weaken the monopoly power of trade unions should be classified as a supply-side policy aimed at reducing the natural rate of unemployment and increasing equilibrium employment and potential output. Such policies would include changes in the law governing trade union activities, or incomes policies — direct regulation of wages — if the aim of the latter is to reduce real wages.'Phis need not he the sole aim
of incomes policies, as we explain in the next chapter. Earlier, we pointed out that frictional and structural unemployment are important com-ponents of the natural rate of unemployment. Policies aimed at reducing frictional and struc-t it ral unemployment should also be included in stipply-side economics. Their objective is to shift he AJ schedule to the right relative to any given position of the labour force schedule LF. Among such policies we include grants that llow redundant workers to retrain in relevant skills, and the various government measures introduced to help school-leavers develop skills and job experience for the first time. By making the labour force more suited to employers' needs, such policies aim to allow firms to make wage offers that unemployed workers will find ceptable. Hence such measures reduce voluntary unemployment.
Policies Aimed at the Demand for Labour Thus far, we have emphasized policies aimed at t he supply of workers for employment. We now turn to the demand for workers by firms. In the previous chapter we saw that an adverse supply s I wk could reduce the demand for labour, shift-ing the LD schedule downwards. It seems plausible that the two dramatic rises in real oil prices in the 1970s had this effect. Overnight, many energy-intensive factories were inade economically obsolete by the rise in oil 1,rices. 'l'hey could no longer compete with more modern energy-saving plant. It was as if many 1st i ng firms had suffered a reduction in their usei HI capital stock. Since labour now had to )1-1K with a smaller quantity of relevant capital equipment, the marginal product of labour was !educed at each employment level. This can be I (presented in Figure 26-5 by a downward shift t be labour demand schedule LD. Try constructing your own diagram to show the effect of this. (Forget about income taxes and %1 .11A Iroiui the point F, in Figure 26-5.) If you draw the diagram correctly you will discover
[more to follow if this chapter is not transcribed yet. It relates to a later post about bad economics teaching in the 1980s which ignored the manufacturing crisis that had caused 3-4 million people to be unemployed, many of them close to the campus, and preferred to teach the usual graphs and a lot of statistics that couldn't be used instead.]
