So a longer delay than I hoped in putting up the rest of the first chapter. Partly that’s because we’ve had visitors, but mainly because I’ve found it tough to whittle down the huge mess of a draft. This was a tough section to write because it summarises later chapters, and it’s hard to find a balance between incomprehensible density and taking up too much space with stuff that will be repeated later. I have ended up leaning towards the dense because all this stuff will be expanded on in the substantive chapters. The density shows especially in the last couple of paragraphs here; it may be better to just cut them and leave to the later chapters. Two more sections still to come, which is going to make this a long chapter – and it may end up being split in two.
A draft thesis section. This is a draft of an unfinished document, please don’t quote without getting in touch first. Quoting in blogs is fine.
Previous section / Next section
December 19, 1969: Friedman's curve shifts upward
Understanding inflation as a policy problem requires an understanding both of its causes – at least as perceived by policymakers – and of the relationships of these causes with policy instruments and other policy goals. If there were a clear chain of influence between a policy instrument and the target of a stable price level, and no competing demands on the instrument, inflation would present no particular problem. The message of the monetarists is that such is the case: the only barrier to price stability is a failure to understand inflation’s nature as “always and everywhere” a question of the money supply (a clear chain of policy influence) and/or that policy can have no long-run effect on unemployment (no competing demands on the instrument). From such a perspective, inflation is a problem only because policymakers misunderstand it – or because they pander to a public that misunderstands it.
In the ‘new macroeconomic consensus’ of the 1990s and 2000s, the monetarist preoccupation with the money supply has been replaced by a focus on the ‘correct’ interest rate. [Arestis, 2007; Arestis and Sawyer, 2008] But the conception of the history of counter-inflation policy – as the eventual triumph of correctness over error – remains. Inflation is explained as a result of what the authorities failed to do. This is the centrepiece of a new crop of neoclassical research into the stagflationary episode of the 1970s. Despite significant debate on the details among this recent literature, Cecchetti et al [2007: 8] note that “[a]ll of these accounts view the Great Inflation as a result of monetary policy error and the Inflation Stabilisation as a restoration of more effective monetary policy.” For example, Nelson  puts forward the ‘monetary policy neglect hypothesis’, while from a different perspective Cecchetti et al [2007: 42] themselves explain ‘the Great Inflation’ in terms of policy deviations from the Taylor rule:
Summing up the international comparisons, three of the four countries exhibit a qualitatively similar pattern in which deviations from a simple policy rule in the 1970s and early 1980s are consistent with the timing of the increases and declines in trend inflation (and its volatility). The peak in the inflation trend and the undershooting of interest rates relative to those implied by a Taylor rule generally occurred around the mid-1970s. There also is some evidence that increases in deviations from policy rules (in an accommodative direction) accompanied increases in trend inflation in the early 1970s.
Yet the Taylor rule – that a central bank should set interest rates according to a formula linking them to the output gap and the distance between actual and target inflation – was not formulated until 1993, as The Economist  dryly notes in reporting on this research. To ‘explain’ the 1970s inflation in this way shows a great deal of confidence that economists have finally worked out how inflation works for once and for all. As we will see, the notion that “the mystery element in monetary policy” [Coombs, 1971 (1954)] has finally been cleared up has been a recurring theme in economic thought. Time and again, paradigms have been knocked over and pre-Enlightenment history re-written as a tragicomedy of grievous, incomprehensible error.
My own story is very different. First, I recognise that inflation has no single cause; rather, it develops from the conjuncture of a number of conditions. It may be argued that this is implicit in the ‘new consensus’ literature with its acknowledgement that ‘potential output’ (i.e., the level of real output associated with a stable rate of inflation) and the corresponding rate of unemployment are not fixed, but depend on factors such as labour productivity, the institutional structure of the labour market, and so on. However, the main point of these models is precisely to fix all these diverse factors in place, at least ‘in the short run’, and in focusing the attention on a few variables, the structure itself is reified and slips into the theoretical unconscious. The apparatus of the output gap, the ‘non-accelerating inflation rate of unemployment’ (NAIRU), and the Taylor rule form an assemblage of a number of factors which could be pulled apart and reassembled in different ways. The particular form it takes represents a decision about which factors to take as given – permanently or in the short run – and which to treat as variables.
Second, I argue that the changing way in which the theoretical structure was assembled not only informed policy but was strongly influenced by the development of policy itself. The material shape of the state policy apparatus within the economic system, and policy strategy in using it, were among the complex of factors determining inflation. For theory, what was considered a constant, what was an exogenous variable, and what was a policy variable depended partly on policy capacity. Or, rather, it depended, on what was perceived as policy capacity, which was subject to dispute. Furthermore, given that price stability was one of a number of policy goals, there was the possibility that policy capacity that could potentially be brought to bear on inflation would not be fully available, given inflation’s interrelationship with other goals. In particular, I argue that inflation theory developed alongside counter-inflation policy in the tension between price stability, full employment and ‘external balance’. Finally, the field on which these tensions played out reflected not only the developing capacities and strategies of policy, but also those of other social actors. All these factors influenced the structure of inflation theory, which in turn informed policy strategy.
So, while my narrative could be read as a long pre-history of the ‘new macroeconomic consensus’ which finally cohered in the 1990s, it is not teleological, while that consensus’s own origin myth is: the consensus was right all along, even before it was formulated, and economists and policymakers eventually realised it. Instead, I present a narrative of a development that could have been different. Policy change comes from contradictions, both internal to policy and external clashes of policy with the defence (and offence) lines of other group-actors. Consequently, also, there is no suggestion in my story that policy history has ended with the consensus: my excavation of the past points to contradictions which still exist below the surface today. (See Chapter 9.)
The non-expectations-augmented Phillips curve relating (inversely) unemployment and inflation is the beginning of the new consensus story, which presents it as the pre-monetarist Keynesian theory of inflation. In my own narrative, it is only the half-way point, already representing a conglomeration of factors which had previously been theoretically separate. It rose to prominence in the 1960s because it appeared to unify two strands of Keynesian inflation theory: ‘demand-pull’ theory focusing on inflation’s relationship with effective demand, and ‘cost-push’ theory centred on its relationship with money-wage growth. These two strands of theory matched separate avenues of policy influence: the first implicated aggregate demand management through fiscal and monetary policy, while the second implicated wages policy. The rise of the Phillips curve internationally was related to the failure of policy to secure direct influence over the money-wage. Targeting it indirectly with aggregate demand put the goal of price stability in conflict with that of full employment, though policy often still aimed to ‘shift the curve’ rather than accept a fixed trade-off. In Australia, where the arbitration system seemed to put the money-wage closer to policy control, the Phillips curve took longer to find policy favour, though a trade-off between unemployment and inflation was recognised as a possibility early on. (See Chapter 4.)
It is far from the case that policy was unaware of the potential for inflationary momentum before the introduction of expectations into Phillips curve models by Phelps  and Friedman . On the contrary, the reaction of money-wages to price inflation was at the centre of Australian policy attention. Even in the Phillips curve literature, there is recognition that experience of inflation could shift the curve – right from Samuelson and Solow’s  original Phillips curve article. The changing way in which inflationary momentum was understood is in itself an interesting story, but as I argue in Chapter 7, is not an explanation for the policy turn of the 1970s. I show that expectations-augmenting the Phillips curve does not explain the 1970s jump in the ‘non-accelerating inflation rate of unemployment’ apparent in such models. The explanation for this jump must be sought elsewhere. I present the policy upheaval of the 1970s and 1980s not in terms of policymakers seeing the light, but as a result of the need to reconcile expectations of a growth rate of real living standards and employment with an economic system that could no longer provide them.
Philip Arestis :”What is the new consensus in macroeconomics?”, in Philip Arestis (ed.), Is There a New Consensus in Macroeconomics?, Palgrave Macmillan, London.
Philip Arestis and Malcolm Sawyer : “A critical reconsideration of the foundations of monetary policy in the new consensus macroeconomics framework”, Cambridge Journal of Economics, 32:5, pp. 761-79.
Stephen Cecchetti, Peter Hooper, Bruce C. Kasman, Kermit L. Schoenholtz and Mark W. Watson : “Understanding the evolving inflation process”, paper presented to U.S. Monetary Policy Forum, available at http://www.brandeis.edu/global/rosenberg_institute/usmpf_2007.pdf, accessed 11 June, 2008.
H. C. Coombs [1971 (1954)]: “The development of monetary policy in Australia”, in Neil Runcie, Australian Monetary and Fiscal Policy: selected readings, v. 1, Hodder and Stoughton, Sydney, pp. 22-43, originally the English, Scottish and Australian Bank Limited Research Lecture, 1954.
The Economist : “Anatomy of a hump”, The Economist, 10 March.
Milton Friedman : “The role of monetary policy”, American Economic Review, 58:1, pp. 1-17.
Edward Nelson : “The Great Inflation of the Seventies: what really happened?”, Working Paper 2004-001, Federal Reserve Bank of St. Louis.
Edmund S. Phelps : “Phillips curves, expectations of inflation and optimal unemployment over time”, Economica, 34: 135, pp. 254-81.
Paul A. Samuelson and Robert M. Solow : “Analytical aspects of anti-inflation policy”, American Economic Review, 50:2, pp. 177-94.