^{(page 86 note 2)} We interpret the CEPG position widely to include ele ments of the thinking and analysis of those who, though not formally members of the CEPG, are closely associated with it—particularly Professor R. R. Neild and Lord Kaldor.

^{(page 87 note 1)} For a discussion see F. T. Blackaby (ed.), De-Industrialisation, Heinemann/NIESR, 1979.

^{(page 88 note 1)} Possible mechanisms are discussed in ‘Pay Bargaining’, National Institute Economic Review, 89, August, 1979. Effects working through expectations may be of varying complexity. The most elaborate hypotheses are based on the idea that economic agents predict the inflation rate on the basis of present and future expected policies and adjust their behaviour accordingly. In monetarist guise, this set of views is known as the rational expectations hypothesis. The general point that anticipations based on policy action may speed up certain adjustments is not contentious.

^{(page 88 note 2)} The term ‘high’ exchange rate is a popular expression in common use but not very precise. Until recently (1979) the only effective mechanism in the LBS model by which inflation was reduced following a restrictive monetary policy was an upward movement in the nominal exchange rate (ceteris paribus). Real international competitiveness was affected during the adjustment. In the most recent versions of the model there are other mechanisms that affect domestic inflation. Strictly, the LBS view on the exchange rate is that it should be allowed to float freely, given an ‘appropriate’ monetary policy.

^{(page 88 note 3)} See for example the submission by T. Burns and A. Budd to the Public Expenditure Committee on EMS. There it is argued that the move to fixed rates would be too sudden a change for the UK. House of Commons, First Report of the Expenditure Committee Session 1977-78. The European Monetary System, London, HMSO, 1978.

^{(page 88 note 4)} The National Institute model does contain an expectations augmented Phillips Curve wage mechanism. In practice, however, this relationship is seldom used for forecasting— wage changes being put in judgmentally or on the basis of other econometric equations. In fact, the National Institute position appears to be one of scepticism about all wage equations.

^{(page 88 note 5)} Supply side problems may arise indirectly. Prolonged recession, by its effect on capital accumulation and productivity, would eventually appear supply constrained.

^{(page 89 note 1)} The Treasury model imposes coefficients for the effect of unemployment and its rate of change on inflation. The ‘Phillips curve’ effect is substantial. In the neighbourhood of 5 per cent unemployment a 1 per cent point increase in unemployment leads in the long run to a lowering of inflation by 0.5 per cent per annum. There is also a short-term effect whilst unemployment is changing. See HM Treasury, Macroeconomic Forecasting Model, Technical Manual 1979.

^{(page 89 note 2)} See for example S. G. B. Henry, M. C. Sawyer and P. Smith, ‘Models of inflation in the UK,’ National Institute Economic Review, 77, August 1976.

^{(page 89 note 3)} In a recent paper Burns and Beenstock present a model (which we understand to be in an experimental stage) which does include pressure of demand effects in both price and wage equations. (M. Beenstock and T. Burns ‘Exchange-Rate Objectives and Macroeconomic Adjustment in the United Kingdom’, in R. L. Major (ed.), Britain's Trade and Exchange-Rate Policy, Heinemann, 1979). Moreover, some versions of the main LBS forecasting model have included such effects. In the case of CEPG some statements suggest that a pressure of demand effect is not ruled out though it is not part of the formal model. See for example R. R. Neild, ‘Managed trade between industrial countries’ in R. L. Major, op. cit.

^{(page 90 note 1)} An extreme version of this thesis is the ‘law of one price’ assumed by monetarists in many of their theoretical writings.

^{(page 90 note 2)} UP until 1979 the LBS relied on this ‘Scandinavian’ mechanism. The most recent version has been modified in two important respects:

• (a) Phillips curve effects are now important whereas previously they had been rejected due to lack of evidence.

• (b) They no longer rely on perfect arbitrage in international trade: mechanisms of the general type used by the LBS are consistent with quite a wide range of price formation processes. See M. Beenstock, The Foreign Exchanges: Theory Modelling and Policy. Ch. 3, Macmillan, 1978.

^{(page 90 note 3)} If, as is likely, productivity trends are different in traded and non-traded sectors (non-traded sectors being pre-dominantly labour-intensive services) the overall rate of price inflation would reflect this.

^{(page 90 note 4)} Ceteris paribus this would suggest that the costs to industry in the process of adjustment would be increased. Other changes in the model have, however, been introduced, and simulation studies are not available.

^{(page 90 note 5)} Assuming profits cannot be squeezed. The CEPG does not assume that there is an immutably determined real wage. Its model refers, in effect, to the ex ante wage, and ex post wages may diverge from it. If they do, however, inflation tends to accelerate.

^{(page 91 note 1)} If unemployment did increase employers' resistance, or lower the bargaining power of unions, then a level of unemployment could be found which led to no inflation (or no change in the rate of inflation). Monetarists would call this the ‘natural rate’. There is, however, nothing specifically monetarist about such an assumption.

^{(page 91 note 2)} See K. Coutts, R. Tarling and F. Wilkinson, Economic Policy Review 1976. For a forward looking target wage approach see S. G. B. Henry, M. C. Sawyer and P. Smith, op. cit.

^{(page 91 note 3)} Cripps and Godley give the following formula:

r=(1/t) (1- WD/WD*)

r = rate of inflation, t = time between settlements,

WD = ex post real wage, WD* = target real wage.

See T. F. Cripps and W. Godley, ‘Formal analysis of the Cambridge Economic Policy Group Model’, Economica, 1976.

^{(page 91 note 4)} They use a pricing equation which assumes a mark up on normal historical costs. Normal costs lag actual costs. In principle, this degree of freedom is a temporary one. See T. F. Cripps and W. Godley, 1976, op.cit.

^{(page 91 note 5)} The CEPG would regard the possibility of accelerating inflation as realistic and a virtue of their model.

^{(page 91 note 6)} The CEPG simulations do not in fact show this. They show that exchange rate depreciations would, on their model, be accompanied by high inflation. They then argue that the path with high inflation is unacceptable, or not feasible politically.

^{(page 92 note 1)} In fact, it is part of the CEPG case that an increase in output towards potential could have a major effect on wages and/or on international competitiveness. Thus it is some times argued that wages in the UK are not too high in relation to potential output—though they are too high given the present degree of slack.

^{(page 92 note 2)} We do not need to be very precise about the contentious question of how a tight or non-accommodating policy stance would be defined.

^{(page 92 note 3)} There is a methodological difficulty that needs to be brought out. The LBS approach is to demand, ab initio, reasonable ‘equilibrium’ properties from their model. Cambridge, by contrast, are more inclined to base their analysis on empirical relationships that apply to the short term. If their simulation results then show cumulative disequilibria— or ‘impossible’ results, they would then admit that something has to change—without being too specific about what adjust ment that would be. A consequence is that it is difficult for the LBS to show any really disastrous results for the longer term, whilst Cambridge can be accused of being somewhat alarmist.

^{(page 92 note 4)} This needs qualification. It is the ‘ex ante’ real wage that is inflexible. It can be lowered ex post, but at the expense of accelerating inflation.

^{(page 93 note 1)} If this is granted, the hard thing to explain is not their dislike of restrictive policies, but their rejection of a policy of expansion combined with exchange rate depreciation. The answer appears to be that they do like it but regard it as impossible in the short term.

^{(page 93 note 2)} There is a parallel here with one of the more common descriptions of the wage push problem. The joint application of comparability criteria and the ability of individual firms to pay seems bound to lead to inflation of the leapfrogging variety.

^{(page 93 note 3)} A simulation of this kind is presented in M. Posner (ed.), Demand Management, Heinemann/NIESR, 1978. We have not mentioned the development of nominal wages or inflation. These are obviously difficult to forecast and interact with the actual path of the exchange rate. Perhaps one could say that the path of real wages, which depends on the real exchange rate, does not appear in their simulations to present very great difficulties which would imply a marked acceleration of inflation of the cost push variety.

^{(page 93 note 4)} Simulations with the LBS, Treasury and National Institute models are reported in J. Laury, G. Lewis, and P. Ormerod, ‘Properties of macro-economic models of the UK Economy.’ National Institute Economic Review, no. 83, February, 1978. Recent versions of the LBS model have however, somewhat different simulation properties. Burns and Beenstock, op. cit., using their experimental model (which is not the same as the main LBS forecasting model) report the effects of a 5 per cent appreciation. Competitiveness is restored by year 4. In the second year there is a 2 per cent loss as compared with the base run. Simulations with the current LBS forecasting model are not available.

^{(page 94 note 1)} We dodge the problem of defining what a fixed exchange rate is in a period of generalised floating by ignoring it. Those who require precision can consider a fixed effective rate.

^{(page 94 note 2)} In the paper by Burns and Beenstock op.cit. the fitting procedure makes use of a slightly different concept—the natural rate of output—and for some equations the natural rate of world output as well. For a critical account see G. D. N. Worswick in Major (ed.), op.cit.

^{(page 94 note 3)} For an account see Burns and Beenstock, Appendix 1, op.cit. See also J. E. H. Davidson, D. F. Hendry, F. Srba and S. Yeo, ‘Econometric modelling of the aggregate time series relationship between consumers’ expenditure and income in the United Kingdom', Economic Journal, December, 1978.

^{(page 94 note 4)} There is no suggestion that this is the paradigm adopted by the LBS.

^{(page 95 note 1)} As simple a device as the kinked demand curve of the elementary text books is sufficient to show that such behaviour can be consistent with profit maximisation.

^{(page 95 note 2)} Similar wealth effects are embodied in all the major models of the UK. A reduction of inflation (ceteris paribus) is reflationary. Whether an unchanged policy stance (monetary and fiscal) against lower inflation should be regarded as active or passive demand managment is a moot point.

^{(page 95 note 3)} Cf. N. Kaldor, ‘The case for regional policy’. Scottish Journal of Political Economy, 1970.

^{(page 95 note 4)} The gradualism of the LBS contrasts sharply with the position of many Keynesian deflationists who believe that a shock to the system (high unemployment or bankruptcies) is unavoidable if expectations are to be radically altered.

^{(page 96 note 1)} This virtuous circle is suggested in the Labour Govern ment's Green Paper ‘The European Monetary System’. Treasury, Cmnd 7405, HMSO, 1978. For a critique see ‘The European Monetary System’. National Institute Economic Review, no. 87, February, 1979.

^{(page 96 note 2)} This ‘export-led growth’ thesis is well known. See for example: W. Beckerman and associates. The British Economy in 1975, Cambridge University Press, 1965. N. Kaldor, Causes of the Slow Rate of Economic Growth of the UK, Cambridge, 1967.

^{(page 96 note 3)} Until fairly recently the New Cambridge School were prepared to follow the suggestion in N. Kaldor, ‘Conflicts in Policy Objectives’, Economic Journal, 1970, that the economy should be managed by managing the exchange rate.

^{(page 96 note 4)} See, for example, T. F. Cripps and W. A. H. Godley ‘Formal analysis of the Cambridge Economic Policy Group model’, Economica, 1976; W. A. H. Godley and R. M. May ‘Macroeconomic implications of devaluation and import restriction’. Economic Policy Review, 3, 1977; T. F. Cripps ‘Causes of growth and recession in world trade’, Economic Policy Review, 4, 1978; T. F. Cripps and W. A. H. Godley ‘Control of imports as a means to full employment and the expansion of world trade: the UK case’, Cambridge Journal of Economics, 1978; R. R. Neild ‘Managed trade between industrial countries’, in R. L. Major (ed.) op. cit. W. A. H. Godley ‘Britain's chronic recession: can anything be done?’ in W. Beckerman (ed.), Slow Growth in Britain, Oxford, 1979.

^{(page 97 note 1)} The CEPG auction scheme must not be confused with a ‘dual exchange rate’ scheme. The latter is a floating of the exchange rate (ie, a market determined, uniform import tariff cum export subsidy) on a specified category of goods. The former amounts to a market determined tariff on imports without a subsidy to exports.

^{(page 98 note 1)} See Cripps and Godley, 1978, op. cit. If the real wage of the employed is to be kept constant after the imposition of a tariff by disbursement of the tariff revenue alone, the ‘deadweight utility loss’ suffered by them must be assumed to be unimportant.

^{(page 98 note 2)} Thus a devaluation tends to reduce the real wage because the implicit import tax on consumers (mainly workers) is passed on to foreigners (insofar as the terms of trade worsen) or to exporters.

^{(page 98 note 3)} Moreover, in the short period before output expanded and unemployment was reduced there could be a perceived reduction in the standard of living of those in employment as they saw the price of imported manufactures rising before they were aware of the benefits due to reduced taxes, etc.

^{(page 101 note 1)} In M. Posner, (ed.) Demand Management, Heineman/NIESR, 1978 it is reported that Professor Ball was prepared to describe himself as an ‘international Keynesian’. It is worth noting that the appeal of the LBS forecasting model is not limited to those of an international monetarist persuasion, since the principal mechanism by which exchange rate changes feed through to inflation is as consistent with Keynesian views on economic management as with monetarist ones.

^{(page 101 note 2)} See Cripps 1978, op. cit.

^{(page 101 note 3)} See, for example, Economic Policy Review, 1979.

^{(page 101 note 4)} See Cripps and Godley, 1978, op. cit.

^{(page 101 note 5)} See Godley, 1979, op. cit.

^{(page 102 note 1)} For a discussion, see C. J. Allsopp ‘The management of the world economy’ in W. Beckerman (ed.) Slow Growth in Britain, Oxford, 1979.

^{(page 102 note 2)} See Cripps, 1978, op. cit.