The Reform of Macroeconomic Policy

The chapters in this book were written over a period of more than two decades. All except the last three have been published before, and the standpoint of the period in which each of them was written should be borne in mind. No attempt has been made to correct them for the change of date (apart from a few footnotes), though they have been edited somewhat in other respects, in the interests of greater clarity, and with some abbreviation. They thus present a development over time of the author’s ideas on current macroeconomic policy issues, and naturally reflect the particular combinations of policy problems that were in evidence at the time of their writing. The application of the suggested principles of policy in a number of different contexts is intended to make the analysis relevant to a number of different combinations of macroeconomic problems in future periods.

The basic approach to the discussion of macroeconomic policy that continues to be generally applied is that which was useful in the conditions that prevailed in the thirty years or so from the late 1930s to the late 1960s. During those years the principal macroeconomic problem at any given time was either the removal of excess demand or the reduction of unemployment. For the past decade or so, however, the problem of the world economy has generally been to reduce both inflation and unemployment. Yet we continue to try to apply models that are inherently unable to deal with that combination of problems, as they essentially lack one of the dimensions of the dual problem that we are facing. Consequently, policy makers relying on these models often become preoccupied with stopping inflation, which leads them to concentrate their measures disproportionately upon that objective, so that they thereby make unemployment worse. Alternatively, they aim their policy measures solely or primarily at reducing unemployment, and so are likely to make inflation worse.

The theory of macroeconomic policy that has dominated thought and policy making since the later 1930s is essentially one-dimensional in the closed economy (and two-dimensional in the open economy). That is to say, in a closed economy we have been taught to operate on the level of demand — with any or every macroeconomic instrument. When inflation is too rapid the aim has been to use (some or all of) our policy instruments to reduce demand; and when unemployment is too high we have learned to raise the level of demand.

Virtually all the macroeconomic policy discussion in the literature is based on a very limited number of the possible alternative combinations of assumptions about the effects of changes in particular policy instruments on the principal macroeconomic objectives. At the monetarist extreme, budgetary instruments are taken to have no impact on the real variable of unemployment (or real output), whereas monetary policy is taken to influence the price level, with no more than a short-run influence on unemployment. At the other (‘Keynesian’) extreme, all the instruments are taken to have similar effects to one another in reducing unemployment, for any given effect on nominal aggregate demand, an effect that is transferred to the price level as full employment is approached. The possibility of different macroeconomic instruments having different relative effects from one another on the price level and on unemployment respectively has been relatively little discussed. Nor has the possibility that one or more of those instruments may have no effect on one of those objectives, whereas another instrument may affect both objectives.

In Chapter 4, the present writer, with Van Hoa (1987), discussed evidence that different forms of macroeconomic stimulus have different effects on the price level for a given effect on unemployment. The conclusion of that study was that measures of monetary expansion have been by far the most inflationary, and the least likely to have an appreciable downward effect on unemployment of the three forms of stimulus tested. Tax cuts appeared to be followed by a significant fall in unemployment over the immediately ensuing years, but without significant apparent upward effects on inflation. Rises in government outlays were intermediate between these two extremes in appearing not to be followed by significant downward effects on unemployment and by only a fairly small upward effect on inflation.

This chapter brings together empirical evidence from econometric simulations by the OECD, the EEC and the University of Warwick using the four main econometric models of the UK. It uses these results to throw light on the relative effects of different macroeconomic instruments on the current account of the balance of payments.

In a celebrated article in 1945 Colin Clark argued that when taxation reached 25 per cent or more of net national income it tended to cause inflation — not immediately, but over succeeding years.¹ In 1964, and again in a revised form of the same essay in 1970, he returned to the theme and elaborated that argument.² There he illustrated it with figures that he interpreted as implying that the countries with the highest tax ratios tended to have the highest rates of inflation. He pointed to the undoubted fact that inflation had been endemic in the period since the end of the Second World War — a period when the ratios of tax revenue to national income had in virtually all developed countries been well above the 25 per cent figure. More recently, in conversation with the present writer, he confirmed his view on this matter, saying that the positive relationship between high taxes and high inflation, though weak, was a clear one.³ It is probably to be regretted that he originally put such emphasis on the 25 per cent figure — which could easily be contested on both a priori and empirical grounds (though his use of it probably served to draw considerably more attention than it would otherwise have attracted).⁴

The amount of attention given to reducing, or at least holding down, the government’s budget deficit (on some definition or other) in recent years has been at the expense of a rational approach to macroeconomic policy. This chapter outlines the reasons why that approach may have damaging macroeconomic effects. It draws attention to ways in which the costs and benefits of a reduction or increase in government borrowing (‘the budget deficit’) should be considered only in the context of the macroeconomic (and other) effects of the particular combinations of government outlays and revenue with which any change in the budget deficit is brought about.

The intermediate target most widely used in discussions of macroeconomic policy in many countries in recent years been the budget deficit, or in Britain the Public Sector Borrowing Requirement (the PSBR): what its level should be, and whether it should be increased, and if so by how much. This chapter considers a number of the dangers involved in this approach, looking first at some widely known objections (which are, nevertheless, almost completely ignored in most of the public discussion). It then discusses in more detail a particular set of problems that arises out of the fact that different ways of bringing about a given level, or a given change, in the budget deficit (on some definition) can have varying effects on any of the various possible fundamental objectives of macroeconomic policy.

This chapter discusses the relationships between the choice of an appropriate combination of macroeconomic instruments, on the one hand, and the rate of increase in money incomes, especially money wage rates, on the other. It makes no assumptions or assertions about the possibility of achieving any particular form of prices and incomes policies, nor about the costs and benefits of achieving such policies either by direct controls or by agreement. It merely takes the view that, whatever the desirability or feasibility of prices and incomes policies may be, the likelihood of their being successful will be enhanced by an appropriate choice of macroeconomic instruments. It is also argued that any sort of prices and incomes policy is virtually certain to be ineffective if resorted to in the hope of thereby escaping the consequences of using inflationary combinations of macroeconomic instruments.

This chapter considers the implications for the choice of macroeconomic policy instruments of various forms of financial deregulation, and some of the concomitant forms of financial innovation, that have occurred in recent years.

Most economic analysis of macroeconomic policy during and since the Second World War has been carried out in a context of both actual and expected inflation. Indeed, for most of that period it could be said that restraining inflation was the dominant objective of macroeconomic policy. It is true that more recently the restoration of low unemployment has become in many countries (especially in continental Europe and Japan) at least as important an objective as holding down inflation. Moreover, there may be also a third objective. This may be defined as ‘maintaining an adequate but not excessive level of saving or investment at a given level of activity’. This has come to be important in one form or another (often expressed — rather inadequately — in terms of keeping down the current account deficit in the balance of payments, or of holding down the budget deficit, on some definition or other).