MICHAEL STEWART is an economist and prospective Parliamentary candidate for Folkestone and Hythe REX WINSBUR Y is a television producer, past chairman of the Y·oung Fabian Group and prospective Parliamentary candidate for Southend West FABIAN TRACT 350 ') THE FABIAN SOCIETY AND YOUNG FABIAN GROUP 11 Dartmouth Street, London, S.W.l Note.-Tius pamphlet, like all publications of the FABIAN SOCIETY, represents not the 'collective view of the Society but only the views of the individuals who prepared it. The responsibility of the Society is limited to approving the publications which it issues as worthy of consideration within the Labour Movement. October, 1963 • I. Introduction ". N "Incomes Policy" is generally unders·tood to olicy, rbich does something to revent total moneY. incom~ ·sing aster. th total production, and · idel hailed as a way of preventing 'nflaotion and, mdirectly, balance of paY.ments crises and a slow ra.te of growth. We agree that there is no particular sense in a system in which money wages rise much faster than productivity and hence there is a continuous rise in prices. Indeed there are positive disadvantages in such a sys1em. At home rising prices cause certain discontents and difficulties; abroad (unless other countries' pr·ices are rising as well) they cause, under a sy5tem of fixed exchange rtttes, a gradual over-valuation of the currenc and a worsening balance ·of payments. T.here is therefore muc to be said or a policy wfiich prevents-infl'se who question the legitimacy or the practicality of this objeotive we would pose this questi·on: is it 'likely that people can be persuaded n<>t to use the full force of their bargaining power (.in other words, is an anti-inflationary Ine<>mes Policy poss~ble) as long as the inequalities and anomal!ies in the structure of inc<>mes remain as marked as they are today? Our own view is no. In this traot we first of all examine the difficul·ties caused by inflation, and the factors which have led to inflation in the post~war period (Chapters 2 and 3); we then discuss the attempts at an Incomes Policy that have been made since the end of the war (Chapter 4), and consider the implications of an Incomes Policy for the trade union movement and in par.ticular for the present system of collective bargaining (Chapter 5). Af·ter a brief discussion <>f foreign experience (Chapter 6), we attempt to indicate, in Chapter 7, the kind of approach to the problem that we think needs to be made. AN INCOMES POLICY FOR LABOUR 3 2: Inflation-to be Slowed Down or Stopped? O O VER th_e past ten or fifteen years retail prices in Britain have risen, on average, by more than 4 per cent a year. Yet our s•tandard of living has been rising by 2 -3]Jer cent a year. It might theref,ore be asked whether inflation really matters. Is there any reason to suppose that our standard of living would have risen faster if we had had no inflati.on? Will the same rate of inflMion prevent our standard of 1iving from rising as fas.t as it might in the future? · Although a great deal of nonsense is talked about the ev.ils of inflation, the answer to these ques·ti·ons seems to be a qualified "yes". For a decade or so following the last war inflation was regarded as a nuisance, but ·tolerable. For this there were two main reasons. One was that for some years the great economic questrion, particularly in the mind of the publ·ic, was whether the government could mailntain full employment or whether, once the immediate post-war boom was over, there would be a return to the kind of unempioymen•t tbM had character.ised the '.twenties and 'thirties. The achievement of successive post-war governments in maintaining a level of employment that has, ·by pre-war standards, been extremely high, diverted a-ttention from continuous ·inflation, which seemed a smal.l price ·to pay for a continuously booming economy. 'the second reason was the benefit derived from the 1949 devaluation of the pound, which for a while permitted export prices to rise more slowly than home prices without unduly reducing profit margins. Pr.ices in other countries also rose fairly rapidly, though not as fast as ours. These two factors saved us from hav.ing to face the choice between stopping our prices rising, and devaluing again. But by about the middle of the decade our nising price level was beglinning to create difficu'lties both at home and abroad. Domestically, as full employment came to be taken more and more for granted, increasing criticism came to be directed towards continuously r,ising prices, particularly \ \ from those on fixed incomes; and although State benefits were adjusted upwards from time to time to keep broadly in line with rises in the cost of l.iving, no such easy remedy was available for those on fixed private incomes and pensions. At the same. time the balance of payments problem t 1.. became chronic. Tht: history of the late 'fifties and early 'sixties was one of continual balance M payments crises which, while less spectacular than those of 1951 and 1955, were suggestJive of much greater structural weakness: they were not the resul•t of suddenly rocketing import prices or an overheated home economy, but of a failu[e to earn enough on current account which persis.ted year after year. The government attempted to deal with these crises by holding back production in order to aUow some unemployment and spare capacity to develop. Its object was tw<>-fold: to make it more difficult for unions to obtain wage increases, and f,or employers to pass on higher wage-costs ~n higher pr.ices; and to reduce the rate of growth of imports and cause manufacturers to search more aggressively for export markets for the goods they could not sell at home. This policy was not successful. A large volume of output was lost 4 AN INCOMES POLICY FOR LABOUR w~thout any no~iceable improvement tak,ing place in our competitive position. British manufacturers failed to secure an adequate increase in exports,' or to compete more effectively with foreign goods in the home market. This may have reflected price levels already too far out of line with foreign competitors. Nor was the poLicy very successful in preventing further substantial increases in domestic prices. lt became apparent tha·t the limited amount of unemployment and spare capacity that the governrl"fent had created was not sufficient to prevent incomes from rising excessively and hence prices from rising. Any greater degree of unemployment and spare capacity would be politically unacceptruble, and the government drew the inescapable conclusion that inflation can be stopped or slowed down only by attacking the problem at source and ·introducing an Incomes Policy. That a Conservative government should now be attempting a degree of control over the process of fixing wages and profit marg.ins that would have been unthinkable even five years ago is evidence of how seriously it takes the problem. Moreover the problem will probably get more serious. The fact that our balance of payments is fairly satisfactory at the momell't is due less to the success of the present Incomes Policy than to the· substantial improvement in our terms of trade over the past five or six years, and to the fact that many of our overseas competitors have been enjoy:ing boom conditions and their prices have recently been rising as fast or faster than our own. Neither of these factors can be counted on to help us in the future. We are therefore faced with a choice: to introduce a workable Incomes Policy; ~to face ...up_10 the domestiCiiiidT~equences of con~s. It is not impossible to ha·ve continual! rising prices without excessive domestic tscontent and bal ce o men~method could l be devise o tying fixed incomes to the cost of iving so that their real value is maintained in spite of rising prices. Externally, if other countries' prices rose less rapidly than ours, we could periodically devalue, or go over to a floating exchange rate. But in this case if our prices continued to rise faster than other countries, the exchange rate would drift downwards, and the lack of confidence this would engender would be difficult to \ reconcile with operating a reserve currency. M.oreover it would progressively raise the price of imports and hence the cost of living, and increase the pressure for higher wages. Paradoxically, therefore, a floating exchange rate might make an Incomes Policy more rather than less necessary. Nevertheless there are no compelling reasons in fav.our of complete price stability. It is sometimes said that even a slow rate of inflation of the order ·Of one or two per cent a year is dangerous because people come to expect continued inflation, and their attempts to allow for this speed up the rate of inflation in an ever-increasing spiral. There is in fact no theoretical reason to expect thiis, and in practice the experience of this and other countries over the pas·t couple of decades shows that it simply does not happen. The only compelling reason for complete price stability would be complete pric-e stabilit in other countries...::::. and this seems very unlikely to occur. It seems much more likely that different countr•ies w.ill experience different ptli;e trends, with some rls~others stable or eve~ falling; but wtt most o t em ex~nencing a slower degree of inflation --·----.. --- AN INCOMES POLICY . FOR LABOUR than in the late 'forties and 'fifties. These different r.ice tr~ds will call for• occasional changes m exchange rates, from which B-;:Ttain wHl in the Long run probaJbly not escape. But if our prices do not on the whole go up faster than othe[fi"eopre'sthere is no reason t>o expect thatthese changes would need To 5e made freq_~ently, or ~ould _p.ave seriou!_5onsequences. The situation that would make eoonomic policy difficult would be one where B'fitam's....p.ric.es o.o.otinually rose faster than other people's, or were only prevented from doin so by_measures that reduced the rate of growth. Thus it IS not necessary that an Incomes Policy should keep the overall level of ·:prrcescorn:ptelely sta.ble. Perhaps the most that it is reasonable to hope for durin the rest of the 'sixties is to get inflation down to something like one or two per~a year; this sort of figure should not lead to undue problems wi·th the balance of payments (if the exchange ra..te is right in the first place); nor should it e mtolerable to publ.iic opini·on. But to achieve even this will r~the introd~~~io.n of an Incomes Policy .. co"Iisid~rably more successful than...a.ny we...have had s•o far. AN INCOMES POLICY FOR LABOUR 3. What Causes Inflation? I I T often used to be said-and it was sometimes true-that inflation was caused by excessive demand or, in the popular phrase, "too much money chasing too few goods". But .on the whole the pers,istent ·inflation which has been suffered by ourselves and other countries since the war has not been caused by excessive demand. It is true that for short periods demand was excessive and that this did contribute directly to rising prices. In Britain late 1954 to early 1956 was probably such a period. But through aut most of the post-war era pnices did not rise because they were pulled up tby people bidding against each other for an insufficient supply of goods; they rose because unions obtained large increases in money wages and employers pushed their prices up in order to maintain their profit margins on their higher level of costs. That this has been by far the most 'important factor in post-war inflation has been attested to by aU the serious investigal tions of the problem.1 The idea that wage increases have been an independent and substantial cause of the post-war inflation has been very difficult for many people to accept. The essence ·of the matter is that the workings of the economy have been transformed by the advent of full employment. (Although there have been variations in the level of post-war emp1oyment, unemployment has at all times been low compared with the level .of unemployment in the '•thirties ·or in previaus depressions; and for most .of the time law even in comparison with the level of 3 per cent which it was believed at the end of the war would be a major achievement.) Broadly speaking, wages and salaries are fixed by negotiation between employees and employers, either singly or in groups. During the late nineteenth and early twentieth century, when there were periodic depressions, these arrangements led to prices being roughly stable over a long period. Prices at the outbreak of the First World War were not significantly higher than they had been during the Battle of Waterloo. This was because wages not only went up at times of boom; they also came down at times of slump. Thus during the upswing of the trade cycle, when output was rising I I apidly, wages, salaries, and profits would all rise rapidly as well, and the net result tended to be some rise in prices. But when the boom petered out and the economy toppled ·over into a decline, wages, salaries and profits were cut and prices tended to fall. Over a long period the price falls of depression periods tended to cancel out the price rises of boom periods. The maintenance of full employment (subject to mild swings in the level tf~dthis s~ation. There are no prolonged slumps ,during which workers ha~~lower wages in order to remain employed, and employers have to cut prices in a desperate attempt to secure a share' of a diminishing quantity of business. Instead, an individual who 'leaves one job can nearly always find another; unions can demand sub 1 OEEC Report on rising prices, 1961 ; Fourth Report of the Council on Prices. Productivity and Incomes. 1961 ; Evidence to the Joint Economic Committee of the United States Congress, 1960-61. AN INCOMES POLICY FOR LABOUR 7 stantial pay increases without fear of their members being thrown out of work if the increase is obtained; and employers can pass on higher wages in .higher prices, partly because their competitors will also have to pay r higher wages and will ra.ise their prices too; and partly because the govern-1 ment wm, broadly ISpeaking, have to maintain a .level of demand in the \ economy sufficient to let them sell their goods at the higher price-because . otherwise there would be serious unemployment, which the government 1 cannot afford. Just as the incentive for unions to submit wage claims jis greater, the incentive for employers to resist them .is less, since a strike stops pr-oduction and affects profits; a wage increase passed on in higher prices does neither. The trouble with this situation is that there is n.o reason why the total money wage increases gran·ted each year should happen to coincide with \ the increase in total output each year. Of course if all workers received . wage increases in 1ine with the increase in . productiv.ity in their firm or /~ industry this condition would be met. But this would mean that in some industries workers ·would get increases of 10 or 20 per cent a year while ' in others, where productivity was stationary or could not be measured, , they would receive no rise at all. This does not happen, and would obviously not be a tolerable situation. The Pattern of Settlements What in fact has been happening over the past decade .is that each year or so a wage ~ettlement in some industry has tended to establish a pattern which has spread through the economy by way of all sorts of traditional relativities. A settlement on the railways, for example, has tended ~o be followed by similar settlements on the London Underground, and then for London busmen and then for provincial •busmen. In the other direction it has tended to spread to railway workshops and then to the engineering industry. Similarly, traditional differentials between civil servants and university teachers have worked to ensure that if one group gets an increase so does the other-though there may be far better reasons (for example because of recruitment difficulties) in the one case than in the other. In addition to this "annual round" of wage increases there has bee~/ J the phenomenon known as "wage drift". This has been an upward drir-t in earnings quite irrespective of increases secured by industry-wide collective'\' bargaining. Some of this drift has •been caused by increased overtime working, but a great deal of it seems to have been the result -of bidding-up of earnings by employers competing for scarce labour, either directly, by various kinds of plus payments and bonuses, or indirectly, by re-setting piece prices in a way that will enable the worker to earn more. But tqe effects of wage drift have spread far beyond the places where a shortage Of skilled labour has .first caused it to appear. Increases i the earnin s of some groups have triggered off demands for higher pay which can s rea to other areas and otller Jn ustr ies. ometimes tfiese su sequent increase~ aiS'Otake the form of drift, so that earnings rise without any change iJ the collectively-agreed wage rates for the industry as a whole. But in othe cases, particularly in those industries where there is no drift and earning can only rise when the national rate rises, pressure is set up fo.r increases in the national rate. In some cases the causal chain may also run the other 8 AN INCOMES POLIO: FOR ·LABOUR way, and increases in national rates are followed by changes .in the earnings of various groups within the plant in order to restore. the percentage differentials which an increase in the national rate (by effecting a bigger percentage increase .in the earnings of those near< the .. rate than those well above it) has reduced. One o( the most obvious, though least accepted, facts about the post- J war inflation has been that wage increases were not ·secured at the expense of profits. P.rofit margins as a proportion of wage· costs were maintained by the process of raising prices, and apart from year to year fluctuations the ratio of profits to wages and salaries has shown no significant change ove.r the past ten or fifteen years. Workers who secured a large wage increase therefo.re secured .only a temporary advantage, which was wiped out when other workers also secured large wage increases and prices went up in order to maintain profit margins. There is no reason to suppose that wage and salary earners as a whole have done better out ·of large increases than they would have done out of much smaller ones; indeed, as pointed out above, the effect on the rate of growth of the government's attempt to check inflation has been to prevent the standard of living from rising as fast as it might have done. AN INCOM~ POLICY FOR LABOUR 9 4. ·Incomes Policy since th~ War S S EVERAL times since the last. ~ar the problem of the rate of increase in money incomes has pushed itself into the centre of the political stage. There is therefore already a backlog of experience to be drawn on as a guide to future policy on incomes. Much contemporary discussion of the subject takes place in a vacuum, ignoring the lessons, for instance, of the LaJbour Government's "wage freeze" of 1948-50, and Selwyn Uoyd's "pay pause" of 1961. This chapter will examine these and other attempts by the Government to slow down the ·rise in wages. The Labour Government's "wage freeze" That even the Labour Party, in t2£7, had not entirely grasped the necessity of relating incomes policy to the hard facts of .industrial relations is shown by the way in which the wa-ge freeze was broached that year. The Minist«an. In all, a total r of 6 million. -neople were .tberefore either directly or indirectly subject to the Governmentstandstill order. But for privateindustry exhortation was all the QQ_vernment relieowerttwa1wtt=should-not theretore be a diminution of ..Q_ower (which the Labour movement would not want to see anyway) but -preferably an enlargement. At p.fesellt, bi-oa y speaking, the power of the unions is brought to bear within the framework of collective bargaining, augmented by arbitration proce ....!:!Ies. e agreements t at emerge rom ·s ms 1tutional set-up are usually voluntary on ~oth sides, but t ~e a s rare!Y broken. Collective bargaining is therefore the way in which the aspirations of the unions have been harmonised with the rest of the British economic system. Weaknesses in the Present System But let us first of all look..,.aLs..o.me.._of_. the weaknesseS-.in-the present sys~which make it at best a clumsy mechanism for fulfilling its functions. There is a great deal of sham bargaining-what has been called a "Persian market." Even a nodding acquaintance with what goes on round the bargaining table between union and employers' representatives will show that the two sides frequently talk at cross purposes, ignore each other's arguments, and rarely agree even on such apparently simple things as the rise in the cost of living since the last wage increase. The unions enter the Jl,~.tiations with a figme for the wage im:rease._t~~hey~~ want whichJ.s purely a· notional figure-1-at which ba~QM_art. ~.p!g,yers pitch thei counte -offer as low as they dare. U~__the result-is.. sbme sort bat, argues, or fights it our with the employers, and then becomes the trend-se er ToFlbe rest-the rest arguing, with force, that they cannot accept less than the first union, but hoping possibly to get more. Ofiten, tl11s pattern is tied in with the permanent imbalance between public sector an private sector wages which is the main so.urce of disequilibrium in thepresent wage structure. Since engineering workers, for example, usually earn above the nationally negotiated rate and railwaymen do not, railwaymen ire-constantly trying to catch up with the engineers in terms of real earnings; but any success they have is wiped out by the engineers using railway increases as bargaining counters for similar increases for themselves. There is therefore a ratchet effect. But the crucial point about this is the haphazard way in which the general level of a year's round of wage increases is set, and the fact that th~_attern of increases may not me~t either economic or social ne~ A fourth weakness is precisely that the collective bargaining system is inflatipn_ary,_ This is doubly defeating, in that nominal wage increases bear little relation to advances in real spending power, and that inflation has led in the past to the Government deliberately stagnating the economy -a condition in which union power is weakened. Fifthly,___t!;},ere is some evi ence that the ultimate sanction of the unions in the collectiyet;a;: ainin rocess-the strike-is becoming increasingly less usable. There has been no ·or strike since the bus strike of 1958, even 'in tlie pay pause period. Strikes have en threatened, but the failure ? of the engin~ Uriions' ballot on a national strike in 1962 will limit ' their powers-in this respect for some time. ~ Collective Bargaining Collective bargaining, voluntarily carried on, is usually defended on three grounds, by different apologists. lt is argued: 1. That it is the means by which the unions have sought with some success to rectify the imbalance between wage earners and the rest by increasing the share of the national income going to wage earners (this, in a nutshell, is the classic industrial aim of the unions); 2. That the elaborate network of agreements within which collective bargaining takes place, with provision for arbitration and other safety valves, are an intricate mechanism for creating social and economic harmony out of a situation of potential conflict : peace, this argument runs, is of such value that the apparatus should not be tampered with; 3. That the existence of trade unions acting as free bargaining units in the present system (or something similar to it) is integral to a free democratic society. This is a semi-philosophical view which tends to believe that the only alternative to the present sort of system is a totalitarian one in which unions are state organisations whose purpose is merely to convey Government decisions down the industrial line, and mobilise support for them. 16 AN INCOMES POLICY FOR LABOUR All three arguments interlock. All three have to be met if a Labour Government is honestly to face the unions and ask for their support for an incomes policy which, given time to develop the implications inherent m it, will significantly change the unions' mode of operation. I. The share of wages in the national income is remarkably simi1ar in different countries and at different times, and there is no reason to suppose that in advanced industrial countries the existence of unions and collective bargaining, particularly in a full employment economy, has much effect on the distribution of income. The worker's standard of living is affected much more by a rise in the national income (in real terms) than by any change in its distribution. 2. The argument based on peace in industrial relations is a difficult one to prove or disprove. It is, of course, historically true that the growing power of the trade unions bad to be given an institutional framework which incorporated it into the social and economic system of the country. The framework has only reached its present degree of maturity since the last war. It is also true that union leaders regard as most sacrosanct the procedures evolved-breaking procedure is one of the biggest crimes in the unions' book. This is because the establishment of procedure meant by definition the acceptance of the unions as a rightful part of the state; it was, as it were, their official recognition. But it still seems true that what ultimately ensures a state of relative industrial peace is a degree of union power which more or less makes the unions the equal of the employers across the bargaining table. If the employers were to gain preponderant power, the collective bargaining machinery would over the years prove valueless. It is true that since the advent of full employment the unions have been more powerful, in some industries, than the employers. But with the increasing interest of the state in industrial matters, it seems unlikely that the margin of union power will get very wide. As for the arbitration part of the present machinery, perhaps the final answer to this line of argument is that the state is in any case increasingly, and inevitably, becoming the final arbiter. By its general policies it determines the growth rate, and so the growth of real wages. It has, and always has had to some degree, a big say in public sector wages. The Ministry of Labour is frequently called in, and bodies like the Industrial Court and special tribunals of enquiry are basically Government creations. To formalise the role of ultimate arbiter is simply to bring the determining factors out into the open. In any case the present structure is not to be scrapped-the important function of settling disputes on all matters other than wages would remain for it. The crux of any addition to the present structure (which in a formal sense it would be) is whether it gives an acceptable avenue to the legitimate aspirations of the trade unions. If it does this, there is no reason why the present structure of industrial relations, with its admitted disadvantages, should be regarded as sacrosanct. 3. The argument from democracy. The best recent statement of this has been in B. C. Roberts' book "Trade Unions in a Free Society". Mr. Roberts says: AN INCOMES POLICY FOR LABOUR 17 "Free trade unionism is an essential feature of a free society; it is better to leave issues of social conflict to the autonomous solution of the parties rither t a~administqWve or legal re!wlaJi.o.D... Granting freedom to volunTary orgamsahons to settle by collective bargaining such questions as the level of wages clearly removes them from the power of the state." But it is worth noting the consequences of this view, as they are recognised by Roberts. He says : 1. In order to stabilise prices by keeping wage increases within the growth of productivity (which Roberts agrees to be necessary) it is necessary to have "an economic policy which must be framed so as to maintain conditions in which it will not be possible for the unions to obtain wages that exceed the appropriate level, since the employers will not be in a position to pay them ... the essential condition on which price stability and free collective bargaining can be reconciled is an acceptance of the need to use such instruments of economic control as the rate of interest, government spending and taxation." 2. "The problem of an over-rapid rate of increase in money incomes would vanish in some sectors if the general level of employment were not allowed to rise to the point where the demand for 1abour far outstrips the supply. Given a definition of full employment which recognises that price stability requires an average unemployment of 2t per cent, there is a good chance that relative wage and price changes would settle down at levels that did not provoke inflation." The 500,000 unemployed should, Mr. Roberts adds, be adequately recompensed for their se~ The "democratic" theory of collective bargaining is clearly an ex post facto rationalisatlion, as well as iiJ some quarters a. sophisticated defence of the entire economic sta·tus quo. The objections are obvious. The trade uni.ons have always fought the market economy, and rarely see their own bargaining function as part of the essential democratic process (except by comparison with what went before). We all know what reliance on fiscal measures has in practice led to-s·tagnation. The unions are unlikely to accept 500,000 unemployed as a permanent feature of the landscape, and no-one in the Labour movement would suggest they should. But perhaps the oddest feature of Roberts' view is his insistence on an illusory exercise of power by the unions. Let the unions bargain freely, he says, and get what they can-but the Government will then cut their gains down to size, according to its own idea of what wage earners should get. If this is the central feature of free collective bargaining in a modern state, is it not better to find for the unions a way of exercising genuine power, without this elaborate pretence? Surely there is a more constructive way of doing things? This discussion of general ideas about the role of the unions and the value ~f their present method of operation has been necessary because of . the far-reaching implications for union activity of accepting. a Government' incomes policy as a permanent part of the political scene. These implications I will not emerge immediately such a policy is initiated-many of them will appear only after the cautious trial period is over and consolidation sets AN INCOMES POLICY FOR LABOUR in. But it would be foolish to baulk the fact that these implications, however latent, are there. First, it will certainly mean the relegation of the present collective bargaining machinery to a secondary place in the structure since the crucial decision, of the size of the general increase for any given year, will be taken elsewhere. The inherent inertia of institutions may be the biggest obstacle here. But it also means that the central skill of many union leaders, that of a negotiator, will be less important, wjth consequent effects on their general status. But a more important implication is for the structure of the trade union movement. Much more centralisation of power in the bands of the T.U.C. is clearly implied, since the hallowed doctrine of the "sovereign rights" of the individual unions revolves largely, though not entirely, round their claim to individual wage bargaining. Any workable incomes policy would mean that the T.U.C. would have to speak for all, and have at least some power to enforce its word. This coincides with other arguments for reform of the union structure advanced in other contexts. It could be that by changing the externaJ economic context within which the unions work, a Labour Government could encourage that centralisation of power in the unions which is for a variety of reasons desirable. Another implication is the necessary shift in emphasis from purely industrial strength to political strength. Clearly, if the crucial decisions are to be taken at the centre, it is at this semi-political level that power must be brought to bear. It may be that such a shift is only possible under a Labour Government, in whose policy-making the unions have in any case a large say. This supports the contention that the connection between the unions and the Labour Party could be turned to positive advantage to both sides if constructive use were made of it. But it does raise the further question of how far the unions should commit themselves when the democratic process means that a Labour Government may be succeeded by a Tory Government in whose policy-making the unions have virtually no say. The answer, presumably, is that the unions will have the right to withdraw from the machinery, both on a change of Government and, indeed, at any other time. Acceptance of an incomes policy will also have implications for the right to strike. Clearly, to be operable, such a policy cannot have banging over it the threat of a strike by a dissatisfied union. Probably the adjustment here will only come about gradually and voluntarily (as indeed will the whole incomes policy). But the point must be recognised from the outset. Trade Union Consent In face of these complex implications, there is no point in expecting the trade unions •o accept overnight some new piece of wide ranging machinery for operating an incomes policy. They will, rightly, approach it with caution. The Labour Party will have to use its maximum powers of education and persuasion to convince unionists at all levels that it is doing the right thing. Ultimately, incomes policy can only succeed with • the broad consent of the people affected, gained by the democratic process of a party gaining general acceptance of its policy. From this it follows that it AN INCOMES POLICY· FOR LABOUR would be foolish to expect too much, too soon. At the start an incomes policy would have many imperfections and some excessive wage increases may slip through the net. The hope would be that a Government that was consistent in its purpose would eventually be able to establish a more or less fully-fledged system. The advantage of any formal set of rules is that each person knows where he stands, by objective reference : breaches are easier to identify, and injustices easier to avoid. Much the same is surely true of an incomes policy-it is essential that under it everyone should know where he stands in relation to everyone else, and accept that where be stands is justifiable and fair. Further, in any organisation vulnerable to the activity of a dissident minority (a state, a trade union, an incomes poLicy) some degree of coercion or sanction is ~nevitable, in this case to make it at least difficult for one union (as opposed to the unions oollectiveiy) to put an end to the whole scheme. In other words, to achieve ·success an incomes policy must have a more-or-less formal basis, although no-one is suggesting that the unions should immediately enter something entirely rigid, still less something simply dictated by the Government. But a prerequisite of a formal incomes policy is the broad consent of the union movement, and this the Government must win and keep. This chapter therefore has two main implications. One is that if an incomes policy is to be brought in by the next Labour Government, there is much preparatory work to be done by way of reaching an understanding with the unions on its purpose and outline, and this work should be proceeding now. · The other is that a shift from the present hostility of many unions to an incomes policy to a willingness to surrender some of their formal freedom on wages, will probably only be achieved by some sort of deal or quid pro quo in the form of greater union say in central policy making under a Labour Government. If the unions are genuinely involved in planning expansion and discussing the way in which the fruits of expansion should be used, their real power will be enlarged and not diminished. 20 AN INCOMES POLICY FOR LABOUR 6. Incomes Abroad I I N view of the recent interest in the centralised Swedish system of wage bargaining shown by the Trades Union Congress (and others), a brief look at foreign experience is necessary, if only to show that on the whole other countries with more regulated wage bargaining procedures than ours have neither been markedly more successful in combating inflation, nor been operating an "incomes policy" in the sense in which we would define it. In Sweden, as in Denmark, the chief characteristic of the system is that the main wage bargaining takes place on a national scale, between central union and employers' bodies, without Government participation. This la t point is, of course, crucial, since it means that the wage bargain is struck without any direct reference to the wider economic needs of the country. This is not to say that the Swedish system does not have advantages : it does. Industrial relations are peaceful, compared to their disturbed state between the wars; both sides operate with a high degree of discipline; and the existence of centrally negotiated wage contracts for set periods has at least moderated pressure on costs from wages. Moreover the earnings of the lower-paid workers have been raised relatively to other earnings. On the other hand, there are certain inflationary tendencies built into the system; Sweden has bad the same problem as Britain, with wages rising faster than productivity. A major cause of this has been the wage drift of the large proportion of people on piece rates. In between national wage agreements, piece-rate earnings have 'tended tO' move ahead significantly. This has been compensated for in subsequent national wage agreements by giving piece-rate workers little or no increase, and other workers larger than average increases. But in practice this means that the process of national collective bargaining has become a mechanism by which wage advanees obtained by one category, the piece-rate workers, have been spread to all other groups-and it is a process in which the highest rate of wage drift tends to set the pace. So although the ability for self-correction is a valuable feature of the system, it is obtained at the price of creating a distinct earnings-rate spiral. However, the emphasis of Government policy has been on keeping internal price increases down to the level of increases abroad, rather than achieving complete price tability. Prices have in fact tended to rise almost as fa t as elsewhere, so that for all its positive virtues, Sweden cannot be said to have overcome the problem of inflation, even if its effects have not been a acute as in Britain. Much the same can be aid for the Dani b system, under which centralised collective bargaining is supplemented by a State Arbitrator whose concern i imply to find an agreed olution when negotiation have broken down. The Government does not interfere in negotiations, except when the AI'bitrator' effort fail, and the Arbitrator i not expected to concern himself with the national intere t. The inflationary effect of this ha been aggravated by the existence of trong institutional price-wage link , in the form of provi ions for wage increa es if price ri e, o that there ha been a price-wage multiplier magnifying price increa e originating from out ide AN INCOMES POLICY FOR LABOUR 21 the country. Partly as a result, Denmark, like Britain, has suffered from chronic balance of payments troubles. The Australian system has something in common with the Danish, in the sense that it is really a system of compulsory arbitration, and not a method of relating wage increases to overall economic considerations. The Commonwealth Court which adjudicates on cases that cannot be settled by bargaining cannot prevent payment above basic rates in times of labour shortage, and inflationary increases in earnings have resulted. But at least the Court is becoming a forum for discussion of the role ·of wages in the national economy. A system which has some relevant lessons is that of the Netherlands, where there is an explicit wages policy. The Government takes part directly in wage negotiations, as part of economic policy; and there have been conscious attempts to create a wages structure based not on the profitability of the firm or industry, but on the nature of the work performed. In periods of heavy demand, wage increases negotiated in this tripartite (unions, employers, Government) system have been greater than strictly justifiable, mainly due to the familiar pressure from wage drift. But at least one big advantage has resulted from the system. The big wage increases have taken place when the balance of payments was favourable; and conversely, at times of crisis, the Government was able to prevent a price-wage spiral setting in. So the economic policy of the Netherlands has been assisted by the Government's ability to exercise some control over the size and timing of wage increases. AN INCOMES POLICY FOR LABOUR 7. An Approach to a Solution Introduction WTE have seen why it is important that inflation should be slowed down; tl' we have seen what the mechanism of inflation in the post-war period has been; we have briefly described the attempts at an Incomes Policy over the past 15 years; and we have discussed the way in which an Incomes Policy would be bound to affect the functions of the unions. The plain fact is that inflation is not going to be moderated until the money value of wages, salaries and profits increases more slowly than it has in the past. (A faster rate of growth might help a bit, but no possible increase in the rate of growth would bridge more than a small part of the gap, and it is probable that a faster rate of growth would in itself cause money incomes to rise faster than before). Experience has shown that there is no acceptable indirect way of stopping excessive rises in incomes (for example by permitting unemployment to develop in the hope that ·this will make wage increases smaller and profits harder to earn). Therefore the problem must be tackled directly, and at source, by influencing the decisions which result in changes in wages and prices. An incomes policy designed to combat inflation must meet four requirements. It must Moderate increases in wage rates; Reduce the incidence of wage drift; Control higher salaries, directors' fees, etc.; Prevent smaller rises in wage costs being offset by bigger increases in profits. The first of these requirements is, quantitatively, the most important, and at the same time the most difficult one to meet. If it can be met, solution of the other problems should not prove insuperably difficult. For this reason we discuss the issue of wage rates first, and at some length. Wage Rates The real question is, how does one ensure that the average settlement each year is not for 01 6-8 per cent increase (which implies annual inflation of 3 or 4 per cent) but for a 3-4 per cent increase (which implies little or no inflation). It seems to us that there are two quite distinct problems here, and that failure to distinguish between them accounts for much of the muddle into which discussion of .an incomes policy has fallen. The first relates to the size of the average increase obtained by most unions each year or so, the second to the question of relativities. Size of Average Increase A good deal of empirical work has been done1 which suggests that there 1 See for example, A. W. Phillips, The Relationship between Unemployment and the Rate of Change of Money Wage Rates in the U.K., 1861-1957, Economica, November 1958; L. A. Dicks-Mireaux and J. C. R. Dow, The Determinants of (footnote continued on following page) AN INCOMES POLICY · FOR LABOUR 23 IS some association between the level of unemployment and increases in the cost of living on the one hand, and the rate of increase of wages on the other : high unemployment makes for smaller wage increases~rapidly ~~\\.>. ( rising prices for larger ones. But the association is a fairly loose one;. _ although wage increases in a given year will tend to be-hlgher or lower ~· according to what has been happening to unemployment and prices, the ~ actual size of the average wage increase is fairly arbitrary. What seems to happen is that for a large variety of reasons usually too complex to analyse, some group of workers gets a wage rise of, say, 6 per cent, and then, because it is difficult for other unions to settle for less, most other groups of workers negotiate, or are awarded by arbitration, increases of about the same size. It is not always possible to identify the keynote settlement at the time, but when one looks back over the successive "wage rounds" which have taken place over the past 10 or 15 years, the predominant influence of certain settlements stands out very strikingly. And although some of these settlements have clearly been influenced by economic factors such as the level of unemployment and the movement of the cost of living, othocs have equally clearly been influenced by non-economic factors. A notable example of this '?'as provided by the failure of the London bus strike in June, 1958. The busmen returned to work after a seven-week strike without securing any improvement on the increase originally offered. This outcome was seen at the time, probably correctly, as determined by a battle of nerves and public relations between the Minister of Labour and the General Secretary of the Transport and General Workers' Union, and not by the interplay of economic forces. Similarly the small average size of subsequent wage increases in 1958 and much of 1959 seems to have been in large measure due to the psychological climate created by the failure of the bus strike. Jt is true tb_at unemployment was rising throughout 1958, and this might have been expt!fted to moderate wage increases anyway; on the other hand, retail prices rose rapidly during 1957 and the first half of 1958, and this might have been expected to make the unions quite militant in their attempts , to restore the real value of wages. It seems legitimate to infer from this that it may be possible to influencem · the size of keynote settlements, and hence of the succeeding wage round. Obviously a Labour Government would not want this influence to be exercised by the failure of a strike. The only way it could be exercised is by agreement on the part of union executive committees that a reasonable annual wage increase is 3 or 4 per cent rather than anything higher. The conditions under which this agreement could be reached have already been t·ouched on in Chapter 5, and for'm the subject matter of much of the rest of this chapter. But the primary requirement is that the unions 1 should be more directly involved in the formulation of economic policy. Over the past 10 years the unions have been able to claim, with some Wage Inflation : United Kingdom 1946-1956, Journal o~ the Royal St~tistic~l Society Series A (General) Vol. 122, part 2, 1959; R. G. Lipsey, The Relatwnshtp betwee~ Unemployment and the Rate of Change of Money Wage Rates in. the U.K ., 1862-1957, a Further Analysis, Economica, February 1960; and L. A. Dtcks- Mireaux, The Inter-Relationship between Cost and Price Changes, 1946-1959, Oxford Economic Papers, October 1961. 24 AN INCOMES POLICY FOR LABOUR justification, that the Government's economic policy has been too bad, and too much oriented towards other classes in the community, to make it possible for them to co-operate in moderating wage increases. Under a Labour Government the unions must have become sufficiently closely associated with the making of economic policy to be willing to commit themselves to helping to create the conditwns in which the policy can be successful. This means, in particular, that union leaders and members of union executive committees must become convinced that one of the conditions of full employment, a faster rate of growth, stable prices, etc., is an absence of excessive wage increases, and must be willing to explain to their members that wage increases of no more than 3 or 4 per cent are in their own interests. This would not necessarily be as difficult as it may sound. In the first place, a wage increase of 3 or 4 per cent has a prima facie reasonableness about it which an increase of 2! per cent (the Conservative Government's original "guiding light") does not. This may be illogical, but the force of this particular piece of illogicality has been widely attested to. It just seems to be the case that if a man is offered 2 per cent he is certain he is being done down; if he is offered 3 or 4 per cent he is not nearly so sure. Secondly, for the reasons mentioned above, it has been a long time since senior union officials have tried to moderate wage demands. It should not be assumed that if they became convinced that wage increases of more than 3 or 4 per cent did more harm than good, they would inevitably be overruled by their members. Thirdly, union leaders would not be appealing to their members in a vacuum. They would be appealing against the background of other policies (discussed later in this chapter) designed to alleviate some of the factors which are behind large wage demands. Relativities So far we have been discussing the first of the two problems involved in moderating the rise in wage rates-the problem of establishing a figure of 3 or 4 per cent as the normal amount by which the average wage should be expected to rise each year. The second .problem-the question of relarivities -is more difficult. Behind what we have said so far has been the assumption that there is no inherent reason why keynote settlements should not be for 3 or 4 per cent rises rather than for 6 or 8 per cent rises. But settlements can only be kept in the neighbourhood of 3 or 4 per cent by the agreement of the unions themselves-by unions refraining from making full use of their bargaining power as far as wage negotiations are concerned. There is at present no effective way in which one can stop an individual union from securing a larger increase than this, particularly if the industry concerned is prosperous and expanding. Here we really get to the heart of the whole problem of an incomes policy. Workers who feel that their wages are about right in relation to other people's may be persuaded, under the conditions outlined above, to settle for 3-4 per cent rather than 6-8 per cent, provided everyone else does the same. But it is going to be very difficult to persuade workers who feel they are underpaid relative to other workers to refrain from using all their bargaining power in an effort to raise their wages relative to other people's. And, of course, if they do succeed in getting AN INCOMES POLICY FOR LABOUR a much larger increase than other people, the whole post-war process of leap-frogging is likely to start all over again. If one group of workers (A) thinks its wages should be 10 per cent higher than those of another group (B), but group B thinks its wages should be the same as group A's, it is difficult to see bow the essentially voluntary system outlined above can operate : group B will use all its bargaining power to get level with group A; group A will use all its bargaining power to get back ahead of group B. There is only one way out of this : group A and group B must come to some agreement about what their relative wages ought to be. This is not going to happen either easily or quickly. One is not, in fact, concerned with the relative wages of only two groups of workers, though the long failure of London and country busmen to agree on a proper differential between them illustrates the difficulty of getting even two groups to agree on what their relative incomes ought to be. One is concerned with the whole structure of wages, salaries and other incomes over the whole country. In fact the problem is so exceedingly complex that anybody might be forgiven for throwing up their bands and saying that no solution is poss~ble. However the need for an incomes policy is so great-and nowadays so widely accepted-that one must attempt to provide the outline of some kind of solution. We do in fact believe that there are a number of reasons for hoping it may •be possible to reach a greater degree of agreement about what the structure of incomes ought to be than exists at the moment. In the first place, much discontent is caused, not by people feeling that they are paid too little in relation to people doing different jobs, but in relation to people doing much the same kind of job (in terms of skill, responsibility exercised, etc.) in other industries. This applies particularly to people in the public sector, who are often paid less than people doing comparable jobs in private industry; but it is also true of different groups of workers within the private sector. One essential requirement of a successful incomes policy must be a method by which underpaid workers (in the sense of those who earn less than comparable workers elsewhere) can have their pay raised relative to other people's. This must be done by agreement, so that the underpaid workers do not have to exert the full force of their bargaining power in an attempt to get the extra increase, and so that the upgrading of their pay, if they are successful, should not be taken as a signal for other people to try and get extra increases over and above the 3 or 4 per cent "normal" increase. Although it will take time to secure this kind of agreement we hopeit will not prove an insuperable difficulty. The principle of fair comparison has operated for years in many parts of the public sector. Civil servants, doctors, policemen and railwaymen have all been awarded large increases in order to bring their pay up to the same level as people doing comparable jobs in private industry; and by and large these extra increases have been regarded by other groups of workers as restoring fair relativities and not as a reason, in themselves, for further wage demands on their part. It is at least a possibility that it may be feasible to extend the principle of fair comparison throughout the whole economy, so that it becomes increasingly accepted that people doing the same kind of job receive the same kind 26 AN INCOMES POLICY FOR LABOUR of pay, regardless of the industry or sector in which they work. Of course, machinery would be needed to put the principle into practice, and this is touched on later. The more difficult, though related problem, concerns income differentials between different kinds of work. What can one do about the conflict between a skilled and an unskilled man who cannot agree on the size of the skill differentials? Or indeed between any two individuals or groups who each think they should get more than the other? This kind of conflict IS inherent in the whole ethos of our society, and although one may deplore it, one would be foolish to think it can be easily eradicated. But can anything be done about it at all? Socialist Aims Before attempting to answer this question we must be permitted a brief, and mildly visionary, digression. As socialists, we believe in equality. We think that a major long-term aim of the Labour Party must be to promote a much greater degree of equality of wealth and income. Possibly some inequality will always be necessary in the interests of economic efficiency; and possibly we shall all be rich enough one day for some degree of inequality to be too unimportant to bother about. But the present degree of inequality is much too great for socialists to tolerate. A progressive tax system can reduce inequality, and a Labour Government will have to mak~ changes in the present system, which is fairly arbitrary in its impact on income and makes no direct attack on wealth. But a tax system, however progressive, cannot fundamentally alter the existing structure of incomes; it cannot deal with a situation in which people doing similar jobs are paid very different amounts, or in which the incomes of people doing different jobs differ fa~ more widely than can be justified in terms of equity and economic efficiency. Therefore equality can only be achieved. in the long run, by making the structure of pre-tax incomes more equal. We recognise that this is not going to happen quickly-we may be looking as much as a generation ahead. But we do insist that quite apart from the problem of inflation, the Labour Party must have the avowed intention of reducing the inequality of pre-tax incomes, and that this must have implications for the kind of incomes policy it operates during the next five years. We will not criticise it if it moves slowly, but we will criticise it if it moves in the wrong direction. Practical Implications The practical implications of this seem to us to be two-fold. First, the Labour Government must use the tax system to take a substantial further step towards reducing the present inequalities of wealth and post-tax incomes. A wealth tax, levied on personal holdings of wealth in excess of, perhaps, £10,000 is an essential measure, as is higher taxation of dividends, or some form of dividend limitation. Death duties should be made compulsory and not-as is largely the case at the moment-voluntary. Capital gains, which are treated by most people as a form of income, must be taxed as such- whether realised within six months or not. Business expenses must be cracked down on, as must all other methods of paying people in cash or in kind in a way that permits avoidance of tax. By measures such as these the government must make it clear that it really is concerned to reduce the AN INCOMES POLICY . FOR LABOUR 27 major inequalities of wealth and income. Secondly, the government must make it clear that it will only look favour-(1f'l ably on demands for above-average (i.e. above 3 or 4 per cent) increases in V \ wages where certain conditions can be shown to obtain. This is the most difficult part of the whole operation, for it will require the ·co-operation of union leaders to establish the fact that above-average increases except in these special cases will lead directly back to inflation and its probable consequences for full employment and the rate of growth, and will defeat the whole purpose of a policy designed to promote a fairer distribution of income. It will require at least some union leaders to come to feel a ( responsibility towards the whole of the union movement, and not just to their own union. (In fact, it would probably require a number of the most/ influential and respected union leaders to take a step up from their position of union secretary to some higher position, perhaps in the government). One of the conditions permitting an above-average increase would be a t[l persistent, long-term shortage, or prospective shortage, of a particular kind l!1 of labour. (We are, of course, discussing relative shortages, which implies that in other parts of the economy there are relative surpluses. In a fully-employed economy there tend to be absolute labour shortages in many places.) Forecasts of the supply and demand for particular types of labour (of the kind that have been begun by N.E.D.C.) must form a part of the government's planning operations, and steps to secure that enough of each type of labour · is forthcoming will in some cases include above-average increases in pay in some occupations. Generally speaking, these occupations will be those in which incomes are at the moment relatively low. In other cases this will not be so, but there are other perfectly legitimate ways in which young people can be steered towards expanding occupations and away from declining ones, and it would be a mistake to suppose that large inequalities may have to be created or perpetuated in order to secure the "right" distribution of the labour force. In fact one of the main advantages of a vigorous programme to educate and train more people (which is an important objective of the Labour Party) is that it will reduce those inequalities of income which are at present a reflection of the relative scarcity of particular types of labour: people will be less able to command high incomes because of their scarcity value as highly-educated graduates or highly-skilled workers. fj) But the main criteria~ of above-average increases must be the ability to Jl7 show that one is underpaid relative to other people. In the simple case, groups must show that other people doing the same kind of work are getting more. In the more complicated case they must show that differentials exist for which there is no adequate justification in terms of skill, responsibility, arduousness, dirt, danger, etc. The presumption must always be that there should be greater equality unless there are good reasons against it. These general principles will have to be embodied in practical decisions. The first requirement is that the principles should be accepted; there is little point in discussing the details of machinery if no-one is willing to work it. But it can be argued that, s·ince the planning of wages cannot be divorced from the planning of other elements in the economy, guidance over the size and pattern of wage increases should become one of the functions of the N.E.D.C. There will have to be a willingness on the part of the T.U.C. members of N.E.D.C.-and therefore on the part of those they represent 28 AN INCOMES POLICY FOR LABOUR -to take a view on whether particular wage claims are, or are not, reasonable, in the light of the need for a slower rise in total money incomes, greater equality of incomes, and certain changes in the wage structure needed to promote a better distribution of the labour force. There are signs that such •willingness is growing; it must be given every possi·ble encouragement, since without it, little can be done. In order to enable N.E.D.C. to take a view on appropriate wage increases in different sectors of the economy, the N.E.D.C. staff would need to be considerably enlarged, and would have to provide itself with two kinds of information. The first kind wo.,.uld be much more comprehensive information than is available at the moment about the earnings of particular occupational groups throughout the whole economy. This data would cover not only annual earnings, but also such factors as hours worked, length of annual holidays, and other fringe benefits. (If N.E.D.C. was to concern itself with the incomes of those not in employment, as perhaps it should, more information would also be needed about the level and distribution of incomes of old-age pensioners, the unemployed and chronically sick, etc.) Job Evaluation Secondly, an attempt at some broad kind of job evaluation will have to be made. This will obviously be needed in order to permit N.E.D.C. to form a view on a group's claim that its members are being paid less than workers doing comparable jobs elsewhere. It will also be needed as the basis of judgements on whether certain differentials are too big (or too small.) The kind of job evaluation carried out could not, at first, be too elaborate. At the same time, enough would have to be known about such factors as the length of training and degree of skill required in a job, the amount of responsibility carried, the degree of physical arduousness or danger involved, etc., to permit some kind of informed opinion about the justification of a group's claim that its relative pay was too low and that it should get an above-average increase. One requirement of this kind of procedure would be that wage claims should relate to a fairly narrowly-defined class of worker. In cases where one claim covered a large number of very different types of worker-for example the kind of claim that is now put in for three million engineering workers-it might be necessary to break the claim down and consider different groups of workers separately. (This task would be made much easier by a successful simplification of the wage structure of the kind that the unions are now undertaking in the engineering industry.) At the same time, N.E.D.C. could not afford to let itself get bogged down in too much detail. Obviously a single plant may have dozens of what it considers different jobs being done within it; many of the detailed reasons for differences of pay between these jobs would continue to be agreed on, as they are now, within the plant. The idea of using a system of job evaluation as a basis for justifying changes in relativities is not as strange as it may seem. Job evaluation is already the basis for determining the pay of a great many people in the public sector. Industrial civil servants' wages are, broadly speaking, linked to pay in comparable occupations in the private sector; and the Pay Research Unit carries out surveys in the course of which it compares the pay, hours, holidays, etc., of non-industrial civil servants with those of men with similar AN INCOMES POLICY FOR LABOUR 29 qualifications, responsibility, etc. in the universtties, private industry and elsewhere. In many other parts of the public sector pay scales are, implicitly if not explicitly, derived from the pay of appropriate jobs in the private sector. A recent settlement for police pay, for example, was reached on the basis of comparison with the standard rates of a number of skilled occupations in outside industry. The same sort of job comparisons have been made for many other groups in the public sector -for example doctors and dentists, the armed forces and railwaymen (where an enormous amount of comparative evaluation was done at the time of the Guillebaud enquiry). Moreover it should be recognised that a vast amount of job evaluation already goes on in connection with pay claims. Unions advancing claims point to comparable workers who are better paid; employers point to others who get less. Of course a lot of this is time-wasting shadow-boxing, but the point is that a lot of information is available, and only needs to be centrally co-ordinated. Again, it should be recognised that in communist countries there is what amounts to a national system of job evaluation, though this is sometimes associated with big wage differentials. In several of these countries specialisation and multiplication of jobs have gone just as far as they have here. This is not to deny that the move from one system to something more closely approximating the other would be a long and difficult business, but the other system can and does exist. Even in other Western countries, such as Sweden and the Netherlands, job evaluation has come to play a significant role in the mechanism of settling pay claims. Pay in comparable jobs is already considered to have an important bearing on what one's own pay ought to be. We think this feeling is growing, and should be encouraged and made use of by any Incomes Policy introduced by the Labour Party. In so far as an Incomes Policy is successful in slowing down increases in the cost of living, and in so far as government action can reduce inequalities of wealth and unearned income, some of the traditional arguments put forward in support of wage claims will come to lose their force. In these circumstances one might expect that the existing structure of earned incomes would come to have more weight in wage bargaining. The successful working of the system described above would depend in the last resort on the voluntary co-operation of the unions. We think that this co-operation might be forthcoming if a number of key conditions are met: the greater involvement of the unions in the formulation of economic policy; the declared intention of the government to support the raising of the wages of underpaid groups of workers relative to other groups, and of encouraging the elimination of unjustified differentials; an aggressive use of the tax system to reduce inequalities of wealth and income; and measures to control excessive profit margins. But there is a danger that such voluntary co-operation might break down. In particular, it is possible that one powerful union might disagree with N.E.D.C.'s view of the increase it ought to get, and use the full force of its bargaining power, in defiance of the disapproval of the union movement or its leading representatives, to screw a much larger rise out of its employers. If this happened other unions might feel they could not afford to hang back, and the whole edifice of a basically voluntary incomes policy would come crashing to the ground. If one is serious about an Incomes Policy, the logic of the argument leads to the conclusion that the government must, in the last resort, take AN INCOMES POLICY FOR LABOUR powers to disallow excessive increases of this kind. But in practice such powers could only be taken on the initiative of the unions themselves, if they came to feel that a fair and workable incomes policy was being sabotaged by a few excessively self-seeking groups. In this event, the government and the T.U.C. (a more powerful T.U.C.) would have to agree on appropriate sanctions; there is little point in discussing at this stage exactly what those sanctions might be. We have now ·outlined the kind of steps we think might be taken to moderate the rise of wage rates. These methods would also cover most salaries. But certain other incomes could not be dealt with in this way, and must be considered separately. But we must first say something about the problem of wage drift. Wage Drift Wage drift is the increase in earnings that takes place over and above increases arising from changes in nationally negotiated wage rates; the mechanism of wage drift was discussed in Chapter 3'. We doubt whether wage drift can be be entirely eliminated nor, perhaps, is it desirable that it should be. But we think there are two main ways in which its incidence can be reduced. First, the government should put its influence behind a move to abolish piece-rate methods of payment. Quite apart from the difficulties they create for an Incomes Policy (which are illustrated in the case of Sweden where, as was shown in Chapter 6, a very high proportion of workers on piece-rates has the effect of negating central agreement on increases in relative incomes), they are becoming an anachronism in an advanced economy where the amount of output is less and less related to individual eltort; they are an endless source of friction; and they run counter to the whole philosophy of an Incomes Policy based on greater equality.1 As far as employer plus payments are concerned, these reflect, to a con- I siderable extent, persistent shortages of many kinds of skilled labour which it must be the aim of a La·bour Government to eliminate in the interests of a higher rate of growth as well as of an Incomes Policy. The forecasting of manpower needs and the training of far more skilled labour mentioned above will do much to lessen this problem. Some drift of this sort will always exist, and will in some cases be helpful in promoting short-term mobility of labour; but it must and can be much less than at present. Finally, in so far as drift results from adjustments to increases in rates, it should automatically slow down as increases in rates slow down. 1 lt has been argued that piece-rates are necessary for a higher rate of growth. but this is questionable; a faster rate of growth depends very little on individuals working with ever-increasing intensity. It is significa nt that Amercan studies suggest that while the rela.tionship between piece-rates and productivity is exceedingly complex, one important effect is to make the worker much more concerned to impress the rate-setter with the difficulties of the job than to co-operate in raising productivity. It is true that the U.S.S.R. relies heavily on incentive methods ·of payment. but it can be argued that this has not assisted the Soviet rate of growth (which. since the Revolution of 1917. according to the American National Bureau of Economic Research. has been slower than that of the U.S.A) AN INCOM ES POLICY FOR LABOUR 31 Higher Earned Incomes. It will be difficult to control higher salaries, directors' fees, professional earnings, etc., chiefly because they account for such a small proportion of a firm's costs. However we see some grounds for hoping that an Incomes Policy would not be sabotaged by the movement of these higher incomes. First, although the level of these incomes is high, there is no evidence that \ they have been increasing faster than wages and salaries as a whole. If wages and salaries rise more slowly, these incomes should do so too. ·secondly, increases in these incomes should be mitigated by the increase in the number of people coming forward to do these jobs. The very rapid increase in the output of uqiversity graduates over the next decade should eventually have the same effect on high administrative and executive jobs as the increase in the numbers of skilled workers on highly paid manual jobs. However this will take time, and the government must be seen to be tackling the problem of high earned incomes directly. This will require changes in the tax system. Subject to our general view about the need for much greater equality in the income structure, we, do not want to penalise unduly those with high qualifications and great responsibilities. At the same time we think there is more nonsense talked about the need for financial incentives than about almost anything else in the economic calendar. It may not be necessary to increase rates of sur-tax, though the point should be made that high income-earners (other than the very highest) are taxed no more heavily here than in most other countries, and less heavily than in some; but if higher rates of sur-tax did seem necessary in order to get agreement on an Incomes Policy there should be no hesitation in introducing them. But the main attack should be concentrated on the many legal but illegitimate ways in which high income earners avoid taxation. Business expenses are one example; payment by firms for executives' housing, cars, holidays, children's education, etc. is another. The psychological effect of an open and radical attack on these abuses should be considerable. However, the poss