"Tight labour markets, rising commodity prices, and a breakdown of policy credibility explains the acceleration of inflation in the 1970s; the reversal of these conditions explains the disinflation of the 1980s"

Throughout the period of Golden Age growth in the OECD, inflation was running at levels that, from today's perspective, may seem slightly excessive, but were at the time in stark contrast to the sustained inter-war experience of constant or even falling price levels. During the 1960s however, the situation began to deteriorate until inflation became the primary problem that policy-makers were faced with. In the immediate aftermath of the first oil shock of 1973, inflation rose to double-digit figures across almost all OECD countries. Although it settled as the decade progressed, a second burst occurred with the recession of 1979-82. Throughout this period, unemployment was undergoing a sustained increase, with sudden bursts associated with the two recessions of the time; thus the phenomenon of stagflation appeared. The trend was brought under control during the course of the 1980s, to the point where inflation is lower today than at any point since W.W.II. However, unemployment is still high by post-war standards, and there has never been any apparent prospect of an automatic return to the full employment levels of the 1950s, especially given the abandonment of demand-management policies. This period is still an important area of study, however, in order to avoid the prospect of any future acute economic conditions as were seen over this time.

The traditional assumption of a sustainable trade-off between inflation and unemployment had already come under attack from Friedman (1968) before the experience of the 1970s proved that this assumption was unfounded. The unemployment experience over the period shows rising levels towards the late-1960s, yet associated with a rise in prices. Thus "tight" labour markets (in the sense that the full employment would strengthen union bargaining positions, drive up wages, and lead to inflation) could lead to inflation, but once unemployment rose, the price level should fall to compensate. If on the other hand there was no impediment to simultaneous unemployment and inflation, why did they only emerge at this time? Soskice points towards a change in the prevailing ethos of labour relations at this time. Corporatist tendencies had been in evidence in many developed countries, with unions being one leg of a tripartite system involving the government and businesses in devising national economic strategies. The sacrifice that unions were encouraged to make for this arrangement was a moderation of their bargaining claims. However, from about 1969 onwards, this system began to break down, for two reasons. Firstly, wages that had been held down throughout the previous decade began to rise in an attempt to "catch-up" levels of profitability, and secondly, the unusually high profits of the late 1960s prompted increasing restlessness on the part of the rank-and-file members of the unions. This manifested itself in increased militancy, with levels of strike activity increasing rapidly in nearly all European countries. The resolution of this was usually capitulation on the part of governments; higher real wages contributed along with a rise in commodity prices (see below) to a profit squeeze, and together this created a vicious circle whereby real wage costs exceeded productivity levels, and lead to reduced profitability, and thus lower investment (especially in the capital stock), and lower future profits. Soskice also maintains that unions were prepared to sacrifice some labour for high real wages, and thus we can see two of the contributions to the stagflation of the time.

However, some authors are not convinced that there is any link between the levels of unemployment and inflation. In particular, Beckerman and Jenkinson believe that in order to find the reasons for the inflation pattern of the late 1970s and early 1980s, it is necessary to look exclusively at the behaviour of commodity prices in this period. They conclude that there is "no identifiable relationship between aggregate unemployment and wage inflation in most individual industrialised countries." Instead, they maintain that most of the deceleration of inflation in the OECD countries between 1980-82 was due to a fall in commodity prices (following a rise in the 1978-80 period), and not due to higher unemployment. They calculate that on average, 20-25% of the deceleration of import prices for the industrialised countries was due to the rapid fall in commodity prices (of 50-60%) in this two-year period. Furthermore, they believe that the pace of wage inflation can be closely correlated with the prices of primary commodities. The most obvious example of a commodity price rise in the 1970s is the twelve-fold increase in the price of crude oil that occurred in stages in 1973 and 1979-80. It seems difficult to reconcile these facts with the notion of a fall in commodity prices, but Beckerman and Jenkinson believe that in relative terms, there was a small decline in oil prices in the period they are looking at. Less controversial is the experience of primary foodstuffs; always subject to the vagaries of the weather, the failure of the Soviet grain harvest in 1972 and subsequent substitution of it with American oil caused international reserves of wheat to fall dangerously low, thereby prompting a rise in its price. As this situation recovered over the course of the next decade, the price would presumably settle (although this is not explicitly stated). Nonetheless, the authors provide both IMF and UNCTAD figures demonstrating that commodity prices fell between 1980-82 after a rise in 1978-80. Part of this reason could be explained by a rise in several OECD members' currencies, most notably the US dollar and the pound sterling. This would assist these countries as most of their imported commodities come from outside the OECD, but it must be remembered that this would lead to intra-OECD difficulties in the trade of manufactures. The conclusion they draw is that it is the fall in these commodity prices that accounts for almost all of the disinflation of the early 1980s. This view is challenged by Gilbert, who is unconvinced by Beckerman and Jenkinson's data. He calculates that the fall in commodity prices can only be used to explain one-third of the fall in inflation, and then almost exclusively in the US; Gilbert argues that in Europe, commodity prices did not move as rapidly or as with as much magnitude as was previously suggested, and that the impact of the Reagan deficits and the high dollar need to be taken into account. He concludes that the effect of the recession of 1979-82 was larger than many sceptical commentators maintain, although he does not infer that this works through a rise in unemployment, as would generally be expected. However, he fails to elucidate on a third mechanism by which a recession could reduce inflation, and this weakens his criticism. This is not to say that it is not appropriate, for it also seems that to denigrate totally the link between unemployment and inflation, as Beckerman and Jenkinson do, is a bold but ultimately insupportable view. A more positive contribution to the debate from Gilbert would have been helpful in attempting to determine the true impact of the change in commodity prices at this time.

Bruno and Sachs do, however, focus on the impact of unemployment during the post-1973 slowdown. Their focus is on explaining stagflation, and thus is not entirely appropriate for the investigation we are conducting; nevertheless, if one accepts the premise of the NAIRU model of inflation - that unemployment is a requirement to hold down price increases - then in explaining how levels of unemployment rose so rapidly, it can be used to explain policy-makers' reactions to the problems posed by stagflation. They point to the end of the Bretton Woods system of fixed exchange rates as being particularly relevant on the demand side. The subsequent acceleration in nominal money supply growth and associated inflation forced monetary authorities to reduce the growth rate of the money supply, causing a fall in the real money supply in many countries. This rejection of monetary accommodation pushed the economies into recession, causing unemployment to rise. On the supply side, the impact of union militancy was detailed above. Stagflation was caused by an interaction of the excessive real wage growth, excessively expansionary monetary policies during the period 1969-73 (which were compensated for by contractions especially as monetarism became practised in the early 1980s), and commodity price shocks. This provides strong evidence for the forces laid out in the title being important contributory factors towards the pattern of inflation as experienced in the OECD at this time. We have yet to look at the issue of policy credibility, however. In this case, it is assumed that during the periods of high inflation, the governments were at least partly to blame for failing to pursue rigorously enough anti-inflationary policies, not wishing to sacrifice unemployment for price stability. Conversely, when monetary policy is taken out of the government's hands, and is transferred to an independent central bank, the problem of time inconsistency does not arise. However, Posen dismisses this theory, by providing evidence that suggests that the costs of disinflation are no lower in countries without independent central banks than those with. One explanation for this is that the governments in question are "hard-nosed" and are committed to low inflation; thus they are as effective as an independent central bank would be. This way, Posen's evidence wouldn't necessarily invalidate the traditional view, but instead it could be maintained that his evidence is the result of mere historical contingency that has given rise to an indistinguishable commitment to low inflation on the part of whichever authorities were responsible for pursuing price stability in the countries he studied, be they the government or independent officials. Depending on how long this were to last, the traditional view may be reasserted; however, it would take many terms of office for all political parties to assure the relevant economic agents that they were as credible as an independent central bank would be in the pursuit of low inflation.

In summary, it can be broadly maintained that the factors outlined in the title are certainly important in explaining the behaviour of inflation over the period concerned. However, as has been seen, there is much disagreement over how much emphasis should be placed on each factor, or whether additional contributory causes should be sought. However, much of the dynamic that is associated with the process of inflation is still unclear - as Allsop says: "Inflation remains one of the least well understood of economic phenomena." It appears that it has been tamed over the course of the 1990s, but were a resurgence to occur in the future, undoubtedly the lessons learnt from the period under discussion in this piece would provide some answers as to how to minimise the impact on the economy.