Lecture to the memory of Alfred Nobel, December 8, 1977

The Meaning of “Internal Balance”

I. It is a special privilege for me on this occasion to have my name associated with that of Professor Bertil Ohlin. By the younger generation of economists we are no doubt both regarded as what in my country are now known as ‘senior citizens’; but I am just that much younger than Professor Ohlin to have regarded him as one of the already established figures when I was first trying to understand international economics. His great work on International and Interregional Trade opened up new insights into the complex of relationships between factor supplies, costs of movement of products and factors, price relationships, and the actual international trade in products, migration of persons, and flows of capital. Of the two volumes which I later wrote on International Economic Policy – namely, The Balance of Payments and Trade and Welfare – it is in the latter that the influence of this work by Professor Ohlin is most clearly marked.

But Professor Ohlin also made an important contribution to what now might be called the macro-economic aspects of a country’s balance of payments. In 1929 in the Economic Journal he engaged in a famous controversy with Keynes on the problem of transferring payments from one country to another across the foreign exchanges. In this he laid stress upon the income-expenditure effects of the reduced spending power in the paying country and of the increased spending power in the recipient country. In doing so he made use of the usual distinction between a country’s imports and exports; but in addition he emphasised the importance of the less usual distinction between a country’s domestic non-tradeable goods and services and its tradeable, exportable and importable, goods. I made some use of this latter distinction in my Balance of Payments; but looking back I regret that I did not let it play a much more central role in that book.

II. Indeed I realise now, looking back with the advantage of hindsight, that my two books were deficient in many respects. From this rich field of deficiencies I have selected one as the subject for today’s lecture, because it raises an issue which in my opinion is at the present time perhaps the most pressing of all for the maintenance of a decent international economic order.

The basic analysis in The Balance of Payments was conducted in terms of static equilibrium models rather than in terms of dynamic growing or disequilibrium models. The use of this method of comparative statics was a result of Keynes’s work.

Keynes in The General Theory applied Marshall’s short-period analysis to the whole macro-economic system instead of to one single firm or industry. In this model additions to capital stocks are taking place; but we deal with a period of time over which the addition to the stock bears a negligible ratio to the total existing stock. Variable factors, and in particular labour, are applied to this stock with a rising marginal cost until marginal cost is equal to selling price, – an assumption which can be modified to accommodate micro-economic theories about determinants of output and prices in conditions of imperfect competition. The rest of the Keynesian analysis with its consumption function, liquidity preference, and investment function can be used to determine the short-period, static, stable equilibrium levels of total national income, output, employment, interest, and so on, in terms of such parameters as the money wage rate, the supply of money, entrepreneur’s expectations, rates of tax, levels of government expenditure, and the foreign demand for the country’s exports. The model can then be used to show how changes in these parameters would affect the short-period equilibrium levels of the various macro-economic variables. Keynes was not, I think, interested in the process of change from one short-period equilibrium to another, though he was very interested in the way in which expectations in a milieu of uncertainty would affect the short-period equilibrium, in particular through their effect upon investment. If my interpretation is correct, he judged intuitively that the short-period mechanisms of adjustment were in fact such that at any one time the macro-elements of the system would not be far different from their short-period equilibrium values; and he may well have been correct in this judgement in the 1930’s.

The Balance of Payments was essentially based on macro-economic models of this kind. What I tried to elaborate was the international interplay between a number of national economies of this Keynesian type. For this purpose I discussed the different combinations of policy variables which would serve to reconcile what I called ‘external balance’ with what I called ‘internal balance’. By ‘external balance’ was meant a balance in the country’s international payments; and although this idea presents, and indeed at the time was realised to present, considerable conceptual difficulties, nevertheless I still instinctively feel that it is not a foolish one. But can the same be said of the idea of ‘internal balance’? Does it mean full employment or does it mean price stability?

I don’t believe that I was quite so stupid as not to realise that full employment and price stability are two quite different things. But one treated them under the same single umbrella of ‘internal balance’ because of a belief or an assumption that if one maintained a level of effective demand which preserved full employment one would also find that the money price level was reasonably stable. The reason for making this tacit or open assumption was, of course, due to a tacit or open assumption that the money wage rate was normally either constant or at least very slugghish in its movements. In this case with the Keynesian model the absolute level of money prices would be rather higher or lower according as the level of effective demand moved the economy to a higher or lower point on the upward-sloping short-period marginal cost curve. But there would be no reason to expect a rapidly rising or falling general level of money prices in any given short-period equilibrium position.

This may have been a very sensible assumption to make in the 1930’s. It is more doubtful whether it was a sensible assumption to make in the immediate post-war years when The Balance of Payments was being written. In any case if I were now rewriting that book I would do the underlying analysis not in terms of the reconciliation of the two objectives of external balance and internal balance, but in terms of the reconciliation of the three objectives of equilibrium in the balance of payments, full employment, and price stability.

Why did I not proceed in this way in the first place?

I was certainly aware of the danger that trade union and other wage-fixing institutions might not permit the maintenance of full employment without a money cost-price inflation. But I suppose that writing immediately after the war I adopted the basic model which was so useful before the war and simply hoped that somehow or another it would be possible to avoid full employment leading to a wage-price inflation. Having done so I found that there remained quite enough important international relationships to examine even on that simplifying assumption. That is not perhaps a very strong defence of my position, but I suspect that it is the truth of the matter.

I am well aware that I could now adopt a more sophisticated line of defence of my past behaviour. It is quite possible to define as the natural level of employment, that level which – given the existing relevant institutions affecting wage-fixing arrangements – would lead to a demand for real wage rates rising at a rate equal to the rate of increase of labour productivity. One has only to add to this the assumption that one starts from a position in which there is no general expectation of future inflation or deflation of money prices to reach the position in which the maintenance of this natural level of employment is compatible with price stability. If this natural level of employment is treated as ‘full employment’ one has succeeded in defining a situation of ‘internal balance’ in which ‘full employment’ and ‘price stability’ can be simultaneously achieved

One could then go on to discuss the many institutions which affect this so-called full employment level. Decent support of the living standards of those who are out of work may mean that unemployed persons are legitimately rather more choosey about taking the first alternative job which is offered to them, quite apart from the existence of a limited number of confirmed ‘sturdy beggars’ who prefer living on social benefits to an honest day’s work. The obligation to make compulsory severance or redundancy payments when employees are dismissed may make some employers less willing to expand their labour force in conditions in which future developments are uncertain. Monopolistic trade union action may put an extra upward pressure on money wage demands which means that unemployment must be maintained at a higher level in order to exert an equivalent countervailing downward pressure. Some statutory wage-fixing bodies in particular occupations may exert a similar influence.

It is not very helpful to squabble about definitions. There is, however, a very real difference of substance between those who do, and those who do not, consider these wage-fixing institutional arrangements to cause very real difficulties. Is it necessary to achieve some radical reform of these institutions in order to make reasonable price stability compatible with reasonably low levels of unemployment? Or is it a fact that, if affairs could for a time be so conducted as to remove the expectation of any marked future inflation of the money cost of living, we would find that even with present institutions the natural level of unemployment would not be at all excessive? I myself would expect that in many countries including the United Kingdom the recasting of wage-fixing arrangements would still be found to be of crucial importance.

As far as the less important question of definition is concerned, I prefer to think of ‘full employment’ and ‘price stability’ as being two separate and often conflicting objectives of macro-economic policies. Anyone who has this preference can, of course, be legitimately challenged to define what is meant by full employment. Perhaps I would be driven to the extreme of defining full employment as that level of employment at which the supply-demand conditions would not lead to attempts to push up the real wage rate more rapidly than the rate of increase in labour productivity if there were perfect competition in the labour market – no monopsonistic employers, no monopolistic trade unions, no social benefits to the unemployed, no obligations on employers to make compulsory severance or redundancy payments to dismissed workers, and so on, – though I am not at all sure whether this extreme form of definition has much meaning. However, in so far as full employment could be defined somewhere along these lines, one would end up with price stability and full employment as separate macro-economic objectives in any real world situation with wage-fixing arrangements as one of the instruments of policy. This is the way in which I like to think of macro-economic problems.

If one adopted this approach, how should The Balance of Payments be recast? In the basic model we would have the three targets of external balance, full employment, and price stability. If one continued to think in terms of matching to each ‘target’ a relevant policy ‘weapon’, one could divide the weapons into three main armouries: the first containing the weapons which directly affect the level of money demands (e.g. monetary and budgetary policies); the second containing the weapons which directly affect the fixing of money wage rates; and the third containing the weapons which directly affect the foreign exchanges, such as the fixing of rates of exchange, measures of exchange control, and commercial policy measures designed directly to affect the total value of imports and exports.

My subsequent education in the rudiments of the theory of control of dynamic systems suggested to me that this was not the best way to have proceeded. One should not pair each particular weapon off with a particular target as its partner, using weapon A to hit target A, weapon B to hit target B, and so on. Rather one should seek to discover what pattern of combination of simultaneous use of all available weapons would produce the most preferred pattern of combination of simultaneous hits on all the desirable targets. With this way of looking at things no particular weapon is concentrated on any particular target; it is the joint effect of all the weapons on all the targets which is relevant.

There is no doubt that this is the way in which a control engineer will look at the problem and that in a technical sense it is the correct way to find the most preferred pattern of hits on a number of targets simultaneously. For a considerable period between the writing of The Balance of Payments and the present time I was fully enamoured of this method.

But I am in the process of having second thoughts and of asking myself whether the idea of trying to hit each particular target by use of a particular weapon or clearly defined single armoury of weapons is really to be ruled out. This onset of second childhood is due to a consideration of the political conditions in which economic policies must be operated. It is most desirable in a modern democratic community that the ordinary man or woman in the street should as far as possible realise what is going on, with responsibilities for success or failure in the different fields of endeavour being dispersed but clearly defined and allocated. To treat the whole of macro-economic control as a single subject for the mysterious art of the control engineer is likely to appear at the best magical and at the worst totally arbitrary and unacceptable to the ordinary citizen. To put each clearly defined weapon or armoury of weapons in the charge of one particular authority or set of decision makers with the responsibility of hitting as nearly as possible one well defined target is a much more intelligible arrangement.

Of course there are obvious disadvantages in any such proposal. Thus the best way for authority A to use weapon A to achieve objective A will undoubtedly be affected by what authorities B and C are doing with weapons B and C. It depends upon the structure of relationships within the economic system how far these repercussions are of major importance. Perhaps a mysterious dynamic model operated inconspicuously in some back room by control experts for silent information of the authorities concerned might be useful; and in any case in the real world it would be desirable for the different authorities at least to communicate their plans to each other so that, by what one hopes would be a convergent process of mutual accommodation, some account could be taken of their interaction. But in the modern community there is, I think, merit in arrangements in which each authority or set of decision makers has a clear ultimate responsibility for success or failure in the attainment of a clearly defined objective.

III. There are six ways in which each of three weapons can be separately aimed at each of three targets. Some of these patterns make more sense than others. In this lecture I can do no more than give a brief account of that particular pattern which, as it seems to me at present, would make the best sense if one takes into account both economic effectiveness and also comprehensibility of responsibilities in a free democratic society. With this pattern

1) the instruments of demand management, fiscal and monetary, would be used so to control total money expenditures as to prevent excessive inflations or deflations of total money incomes;

2) wage-fixing institutions would be modelled so as to restrain upward movements of money wage rates in those particular sectors where there were no shortages of manpower and to allow upward movements where these were needed to attract or retain labour to meet actual or expected manpower shortages, thus preserving full employment with some moderate average rise in money wage rates in conditions in which demand management policies were ensuring a steadily growing money demand for labour as a whole; and

3) foreign exchange policies would be used to keep the balance of payments in equilibrium.

This pattern implies the use of the weapons of demand management to restrain monetary inflation and of wage-fixing to influence the real level of employment and output. Many of my friends and colleagues who share my admiration for Keynes will at this point part company from me. ‘Surely’, they will say, ‘you have got it the wrong way round. Did not Keynes suggest that the control of demand should be used to influence the total amount of real output and employment which it was profitable to maintain, while the money wage rate was left simply to determine the absolute level of money prices and costs at which this level of real activity would take place?’ I agree that this is in fact the way in which Keynes looked at things in the late 1930’s when it could be assumed that the money wage rate was in any case rather constant or sluggish in its movements. What he would be saying today is anybody’s guess; and I do not propose to take part in that guessing game except to say that he would be appalled at the current rates of price inflation. It is a complete misrepresentation of the views of a great and wise man to suggest that in present conditions he would have been concerned only with the maintenance of full employment and not at all with the avoidance of money price and wage inflation.

But, whatever Keynes’ policy recommendations would be in present circumstances, I would maintain that the way in which I have distributed the weapons among the targets is in no way incompatible with Keynes’ analysis. In the 1930’s Keynes argued rightly or wrongly, that cutting money wage rates would have little effect in expanding employment because its main effect would be simply to reduce the absolute level of the relevant money prices, money costs, money incomes, and money expenditures, leaving the levels of real output and employment much unchanged. It is a totally different matter, wholly consistent with that Keynesian analysis, to suggest that the money wage rate might be used to influence the level of employment in conditions in which the money demand was being successfully managed in such a way as to prevent changes in wage rates from causing any offsetting rise or fall in total money incomes and expenditures. If one is going to aim particular weapons at particular targets in the interests of democratic understanding and responsibility, it is, in my opinion, most appropriate that the Central Bank which creates money and the Treasury which pours it out should be responsible for preventing monetary inflations and deflations, while those who fix the wage rates in various sectors of the economy should take responsibility for the effect of their action on the resulting levels of employment.

Earlier I spoke of ‘price stability’ as being one of the components of ‘internal balance’. Yet in the outline which I have just given of a possible distribution of responsibilities no one is directly responsible for price stability. To make price stability itself the objective of demand management would be very dangerous. If there were an upward pressure on prices because the prices of imports had risen or because indirect taxes had been raised, the maintenance of price stability would require an offsetting absolute reduction in domestic money wage costs; and who knows what levels of depression and unemployment it might be necessary consciously to engineer in order to achieve such a result? This particular danger might be avoided by choice of a price index for stabilisation which excluded both indirect taxes and the price of imports; but even so, the stabilisation of such a price index would be very dangerous. If any remodelled wage-fixing arrangements were not working perfectly, – and it would be foolhardy to assume a perfect performance – a very moderate excessive upward pressure on money wage rates and so on costs might cause a very great reduction in output and employment if there were no rise in selling prices so that the whole of the impact of the increased money costs was taken on profit margins. If, however, it was total money incomes which were stabilized, a much more moderate decline in employment combined with a moderate rise in prices would serve to maintain the uninflated total of money incomes.

The effectiveness of the pattern of responsibilities which I have outlined rests upon the assumption that there is a reasonably high elasticity of demand for labour in terms of the real wage rate, since success is to be achieved by setting a money wage rate relatively to money demand prices which gives a full employment demand for labour by employers. I have no doubt myself that in the longer-run the elasticity of demand is great enough. But what of the short-run? What if in every industry there is a stock of fixed capital in a form which sets an absolute limit to the amount of labour which can be usefully employed, while for some reason or another of past history there is more labour seeking work than can be usefully employed? There will be unemployment in every industry; and any resulting reduction in money wage rates combined with the maintenance of total money incomes would merely redistribute income from wages to profits.

I have explained the danger in its most exagerated form; but it would remain a real one even in a much moderated form. There should, of course, never be any question of the wholesale immediate slashing of wage rates in every sector in which there was any unemployment. Any such arrangement would, for the reasons which I have outlined, be economically most undesirable even if it were politically possible. What one has in mind is simply that in a milieu in which total money incomes are steadily rising at a moderate rate, money wage rates should be rising rather more rapidly in some sectors and less rapidly or not at all in other sectors according to the supplies of available labour and the prospects of future demands for labour in those sectors. There would be no requirement that any money wage rates must be actually reduced.

But putting more emphasis on supply-demand conditions in the settlement of particular wage claims could only work if there were general acceptance of the idea by the ordinary citizen; and such acceptance would depend inter alia upon a marked change of emphasis about policies for influencing the redistribution of the national income. I have for long believed that it is only if, somehow or another, the ordinary citizen can be persuaded to put less emphasis on wage bargaining and more emphasis on fiscal policies of taxation and social security for influencing the personal distribution of income and wealth that we have any hope of building the sort of free, efficient, and humanely just society in which I would like to live. But that raises a host of issues which I can not discuss today.

There is, however, one feature of this connection between the supply-demand criterion for fixing wage rates and the attitude of the wage earner to his real standard of living on which I do wish to comment. Suppose, for example because of a rise in the world price of oil or of other imported foodstuffs or raw materials, that the international terms of trade turn against an industrialised country. This is equivalent to a reduction in the productivity of labour and of other factors employed in the country in question. If money wage rates are pushed up as the prices of imported goods go up in order to preserve the real purchasing power of wage incomes, money wage costs are raised for the domestic producer without any automatic rise in the selling price of the domestic components of their outputs. Profit margins are squeezed. The demand for labour will fall unless and until profit margins are restored by a corresponding rise in the selling prices of domestic products. But such a rise would in turn cause a further rise in the cost of living, followed perhaps by a further offsetting rise in money wage rates, with a further round of pressure on profit margins. In fact workers are attempting to establish a real wage rate which, because of the adverse effect of the terms of international trade, is no longer compatible with full employment. The resulting rounds of pressure on profit margins, rises in domestic selling prices, further rises in money wage rates, further pressure on profit margins, and so on, may result in stagflation – a level of employment below full employment with a continuing inflation of money prices.

This story may in fact help to explain what has happened recently in some industrialised countries. But my purpose in telling it is merely to give a vivid illustration of the fact that an effective combination of full employment with the avoidance of inflation necessarily requires that wage-fixing should take as its main criterion the supply-demand conditions in the labour market without undue insistence on the attainment and defence of any particular real wage income. The latter must be the combined result of domestic productivity, the terms of international trade, and tax and other measures taken to affect the distribution of income between net-of-tax spendable wages and other net-of-tax incomes.

IV. So much for the specification of targets and for the distribution of weapons among targets. What about the detailed specification of the weapons themselves?

If the velocity of circulation of money were constant, a steady rate of growth in the total money demand for goods and services could be achieved by a steady rate of growth in the supply of money, and this in turn could be the task of an independent Central Bank with the express responsibility for ensuring a steady rate of growth of the money supply of, say, 5 per cent per annum. It is a most attractive and straight forward solution; but, alas, I am still not persuaded to be an out-and-out monetarist of this kind. It is difficult to define precisely what is to be treated as money in a modern economy. At the borderline of the definition substitutes for money can and do readily increase and decrease in amount and within the borders of the definition velocities of circulation can and do change substantially. Can we not use monetary policy more directly for the attainment of the objective of a steady rate of growth of, say, 5 per cent per annum in total money incomes, and supplement this monetary policy with some form of fiscal regulator in order to achieve a more prompt and effective response? For this purpose one would, of course, be well advised to call in aid the skills of the control engineer in order to cope with the dynamic problem of keeping the total national money income on its target part. Am I to be regarded as a member of the lunatic fringe or as an unconscious ally of authoritarian tyranny if I express this remaining degree of belief in the possibilities of rational social engineering?

I find very attractive the idea that this monetary control should be the responsibility of some body which was not directly dependent upon the government for its day-to-day decisions but which was charged by its constitution independently to achieve this stable but moderate growth of money incomes. But there is real difficulty in endowing any such independent body with powers to use fiscal policy as well as monetary policy to achieve its objective.

Let me take an example. Suppose that overseas producers of oil raise abruptly the price charged to an importing country; and, to isolate the point which I want to make, suppose further that the oil producers invest in the importing country any excess funds which they receive from the sale of an unchanged supply of oil, so that there is no immediate need to cut imports or to expand exports in order to protect the foreign exchanges. The abrupt price change will, however, tend to cause a deflation of money incomes in the importing country whose citizens will, out of any given income, spend less on home produced goods in order to spend more on imported oil, the receipts from which are saved by the oil producers. With the scheme of responsibilities which I have outlined it is now the duty of the demand managers to reflate the demand for goods and services in the importing country in order to prevent a fall in money incomes in that country.

But there are at least two alternative strategies for such reflation.

In the first place, the taxation of the citizens of the importing country might be reduced so that, while they have to spend more on imported oil, they have just so much more spendable income to maintain their demands for their own products. In this case the government directly or indirectly borrows funds saved by the oil producers to finance the larger budget deficit due to the reduced tax payments by the domestic consumers. No one’s standard of living is immediately affected.

But if this solution is adopted, the importing country faces an ever-growing debt to the foreign oil producers with no corresponding growth of domestic or foreign capital to set against it. If this is considered undesirable, the private citizens must not be relieved of tax; their current consumption standards must be allowed to fall as a result of the rise in the price of oil; and the reflation of domestic incomes must be brought about by measures which stimulate expenditure on extra real capital development at home, the finance of which will mop up the savings of the oil producers. Such action will depend upon monetary policies rather than, or at least as well as, upon fiscal action.

I have told this particular story simply to make the point that the choice between fiscal action and monetary action must often depend upon basic policy issues which should certainly be the responsibility of the government rather than of any independent monetary authority. Perhaps the best compromise is an independent monetary authority charged so to manage the money supply and the market rate of interest as to maintain the growth of total money income on its 5-per-cent-per-annum target part, after taking into account whatever fiscal policies the government may adopt. One would hope, of course, that there would be a suitable discussion of their plans and policies between the government and the monetary authority; but the latter would be given an ultimately independent duty and independent choice of monetary policy for keeping total money incomes on their target path.

The difficulties involved in the specification of the weapons of demand management are real enough; but they fade into insignificance when they are compared with the problems of remodelling wage-fixing arrangements in such a way as to ensure a greater emphasis on supply-demand conditions in each sector of the labour market.

I can think of five broad lines of approach.

First, one can conceive of wage fixing in each sector of the labour market by the edict of some government authority. An efficient use of this method would be extremely difficult in a modern economy with its innumerable different forms and skills of labour in so many different and diverse regions, occupations, and industries. It would, I think, in any case ultimately involve a degree of governmental authoritarian control which I personally would find very distasteful.

Second, there is the corporate state solution in which a monopoly of employer monopolies agrees with a monopoly of labour monopolies on a central bargain for the distribution between wages and profits in the various sectors of the economy of the total national money income which the demand managers are going to provide. I suspect that, in the United Kingdom at any rate, any such bargain would be very difficult to attain without leaving some important, but relatively powerless, sectors out in the cold of unemployment or of very low wages. In any case I ought to reveal my prejudice against being ruled by a monopoly of uncontrolled private monopolists.

Thirdly, the restoration of competitive conditions in the labour market would in theory do the trick, since the competitive search for jobs would restrain the wage rate in any sector in which there was unemployed labour and the competitive search for hands by competitive employers would raise the wage rate in any sector in which manpower was scarce. There is little doubt that in some cases trade unions have attained an excessively privileged position and some reduction of their monopoly powers might help towards a solution. But I do not believe that any full solution is to be found along this competitive road. On the employers’ side it may be impossible to ensure effective competition where economies of large scale severely restrict the number of employers. On the employees’ side the whole of history suggests the powerful psychological need for workers with common concerns to get together in the formation of associations to represent their common interests. Moreover, reliance on individual competition might well involve the reduction, if not elimination, of support for workers who were unemployed and of compulsory severance or redundancy payments to workers whose jobs disappeared. But what one wants to find is some effective, but compassionate and humane, method which applies supply-demand criteria for the fixing of wage rates for those in employment without inflicting needless hardship and anxiety on those particular individuals who are inevitably adversely affected by economic change.

Fourth, there are those who see the solution in the labour-managed economy in which workers hire capital rather than capital hiring workers. In such circumstances there would be no wage rate to fix. Workers would share among themselves whatever income they could earn in their concerns after payment of whatever fixed interest or rent was necessary to hire their instruments of production. These ideas are very attractive; but, alas, there is, I think, good reason to believe that satisfactory outcome on these lines is possible only in those sectors of the economy where small scale enterprises are appropriate and where conditions make it fairly easy to set up new competing co-operative concerns.

Finally, there remains the possibility of the replacement in wage bargaining of the untamed use of monopolistic power through the threat of strikes, lockouts, and similar industrial action by the acceptance of arbitral awards made by trusted and impartial outside tribunals, – awards which would, however, have to be heavily weighted by considerations of the supply – demand conditions of each particular case, if they were to achieve what I have suggested should be the basic objective of wage-fixing arrangements.

This is the civilised approach; but I am under no illusion that it is an easy one. It relies upon a widespread acceptance of the idea that some such approach is necessary for everyone’s ultimate welfare and, in particular, as I have already indicated, upon the belief that there are alternative fiscal and similar policies to ensure social justice in the ultimate distribution of income and wealth. But even if in the course of time such a general acceptance could be achieved, some form of sanction for its application in some particular cases would almost certainly be needed. The punishment of individuals as criminals for taking monopolistic action to disturb a wage award does not hold out much hope of success. But is it pure dreaming to conceive ultimately of a state of affairs in which (1) in the case of any dispute about wages either party to the dispute or the government itself could apply for an award of the kind which I have indicated and (2) certain financial privileges and legal immunities otherwise enjoyed by the parties to a trade dispute would not be available in the case of industrial action taken in defiance of such an award?

Perhaps this is merely an optimist’s utopian fantasy; but I can think of nothing better.

V. So much for the attainment of price stability and full employment through the instrumentalities of demand management and wage-fixing. What about the attainment of external balance through foreign exchange policies?

In my view the appropriate division of powers and obligations between national governments and international institutions is that the national governments should be responsible for national monetary, fiscal, and wage policies which combine full employment with price stability and that external balance should be maintained by foreign exchange policies under the supervision of international institutions.

Variations in the rate of exchange between the national currencies combined with freedom of trade and payments should in my view be the normal instrument of such foreign exchange policies. But this is not to say that there will never be occasion for the use of other instruments of foreign exchange policy. Special control arrangements may be appropriate where the removal of an international imbalance requires wholesale industrial development or structural change, or where abrupt changes in the international flow of capital funds require special offsetting measures, or where differences in national tax regimes would distort international transactions in the absence of offsetting measures. But where such exceptions to the free movement of goods and funds arise, these should be under the rules and supervision of appropriate international institutions.

After the war we managed to lay the foundations of an international system of this kind with the pivotal institutions of the International Monetary Fund, the International Bank for Reconstruction and Development, and the General Agreement on Tariffs and Trade, a system which for a quarter of a century resulted in a most remarkable expansion of international trade. In my opinion there was one important original flaw in this system, namely the insistence on the International Monetary Fund’s very sticky adjustable peg mechanism for the correction of inappropriate exchange rates. But even this flaw has now gone as the International Monetary Fund seeks to find the most appropriate rules for running a system of international flexible exchange rates.

And yet we seem now to be faced with the possibility of a gigantic tragedy, with this initial success being fated unnecessarily to end in calamity. Why is this so? In my view the answer is obvious; it is simply because so many of the national governments of the developed industrialised countries have failed to find appropriate national institutional ways of combining full employment with price stability.

If they could do so, not only would the domestic tragic waste and social discontent of heavy unemployment in such countries be removed, but the international scene would be transformed. The case for the use by developed countries of massive import restrictions rather than of gradual and moderate changes in exchange rates to look after their balances of payments would, I suspect, evaporate. It is the spectacle of imports competing with the products of domestic industries in which there is already serious unemployment which is the greatest threat to the freedom of imports into the developed countries. With full employment and price stability at home the balance of payments could with much more confidence be left to the mechanism of flexible foreign exchange rates. The developed countries would then have less difficulty in giving financial aid to the third world; and, what in my opinion is even more important, they could much more readily accept the inflow from the third world of their labour-intensive products.

In this lecture I have marked an occasion which is concerned with international economics with a lecture on internal balance. But I suggest that in present conditions this is not anomalous. I do not, I think, exaggerate wildly when I conclude by saying that one – though, of course, only one – of the really important factors on which the health of the world now depends is the recasting of wage-fixing arrangements in a limited number of developed countries.