Unit 4 Inflation and unemployment
4.9 Global oil and other commodity price shocks
Before you start
Before beginning work on this section, read Section 2.8, ‘Imported materials and the WS–PS model’, for the derivation of the PS curve including imported materials.
Changes in the global economy can cause supply shocks that trigger inflation. But changes in the climate can do so as well. Confirmation in 2023 of the El Niño conditions (a periodic warming of the equatorial central and eastern Pacific) by the US Climate Prediction Center, led the European Central Bank (ECB) to evaluate the implication of an associated one-degree temperature rise. Previous such events produced a rise in global food prices by more than 6% after one year.
Shocks to the world oil price are one major source of disturbance. A typical firm uses imported oil in the production process. At the level of the economy as a whole, when the country is a net importer of oil, the national pie to be divided between owners and employees shrinks when more has to be paid for imported oil and this heightened conflict over shares can cause a wage–price spiral.
Here is how the German central bank (the Bundesbank) described the problem at the time of the first oil shock in 1973:
The oil-producing countries’ striving for a larger share in the national product and national income of the industrial countries need not necessarily result in a further acceleration of the pace of price rises. Whether this occurs depends in every country very greatly on whether it is made easier or more difficult to pass on the higher prices of these (and other) major imports—in other words, on whether the intensification of the international distribution struggle triggered off by the oil-producing countries’ price agreement is followed by an intensification of the domestic struggle for the distribution of the national income… The aim of the Bundesbank’s policy, in full agreement with the Federal Government, is to restrict the scope for passing on the higher prices.1
Following the early 1970s oil shock, US inflation jumped from 6.2% in 1973 to 9.1% in 1975 and unemployment went from 4.9% to 8.5% at the same time. This pattern was common across the high-income world. For example, in the same period, inflation in Spain rose from 11.4% to 17% and unemployment increased from 2.7% to 4.7%.
Figure 4.22 shows that there were two big recessions in the UK in the 1970s. They followed the oil shocks of 1973–1974 and 1979–1980, which were associated with a rise in both unemployment and inflation to their highest levels since the Second World War. Unemployment in the UK peaked at nearly 12% in the mid-1980s.
Figure 4.22 UK GDP growth and real oil prices (1950–2022).
UK Office for National Statistics. 2023; Ryland Thomas and Nicholas Dimsdale. 2017. ‘A Millennium of UK Data’; Bank of England OBRA dataset; Federal Reserve Bank of St. Louis. 2023. FRED.
Figure 4.23 UK inflation, equilibrium unemployment rate (‘NAIRU’), and real oil prices (1971–2022).
UK Office for National Statistics. 2023; Ryland Thomas and Nicholas Dimsdale. 2017. ‘A Millennium of UK Data’; Federal Reserve Bank of St. Louis. 2023. FRED. OECD. Equilibrium unemployment rate. Accessed October 2023.
Figure 4.23 presents an estimate of the equilibrium or structural unemployment rate. This is labelled the ‘NAIRU’, which stands for non-accelerating inflation rate of unemployment. In the WS–PS model, we use the term equilibrium or structural unemployment for the rate of unemployment at which inflation is constant. When economists use data to estimate this concept, it is called the NAIRU. This is a much smoother series than the annual unemployment rate. It captures changes on the supply-side of the economy.
Remember that when unemployment is at its equilibrium level, inflation is stable: it is constant and equal to expected inflation. And at lower rates of unemployment, inflation rises. So it would be more accurate to refer to equilibrium unemployment as ‘the non-rising inflation rate of unemployment’ or ‘inflation-stabilizing rate of unemployment’. But the term ‘NAIRU’ has stuck—so be careful not to be misled by it.
In the UK economy, the structural unemployment rate doubled from 5% to 10% between 1970 and 1985. One of the factors driving that increase was oil prices.
The WS–PS model and the Phillips curve can explain why a one-off increase in the world oil price can lead to a combination of:
- higher structural unemployment
- a rise in the price level (inflation) at the time of the shock, and
- rising inflation.
A rise in the oil price shifts the price-setting curve down: this leads to higher structural unemployment, a positive bargaining gap at the pre-existing level of employment, and inflation. The Phillips curve will continue to shift up as expected inflation rises.
When we allow for imported input materials, the marginal cost is higher than in the model where labour is the only input. Then the price will also be higher by the same factor, and the price-setting real wage will be lower. If imported materials raise the marginal cost by a proportion, τ, the price-setting wage is:
WP=(1−σ)λ(1+τ)and we can model the effect of a rise in imported materials prices as an increase in τ, lowering the price-setting real wage. The PS curve shifts down, because firms pass on the higher costs in the price of their products. Real output per worker is shared between foreign suppliers of input materials, workers, and firms.
Once again, as long as the conditions for competition in their markets for goods and services are unchanged, firms continue to get the markup, σ, and an increase in the price of imported materials leaves the profit share unaffected. With the same amount of output per worker going to owners and more going to foreigners supplying inputs, there is less for the real wage. An increase in energy prices, for example, means that real wages fall as shown by the downward shift of the PS curve, and the structural unemployment rate at supply-side equilibrium rises.
In more detail, firms raise their prices to protect their profit margins when the cost of imported oil rises. This reduces the real wage of employees, so the price-setting curve shifts down. At the initial employment level, this opens up a positive bargaining gap between the lower real wage on the price-setting curve and the unchanged real wage on the wage-setting curve. Firms are able to protect themselves from the impact of the energy shock by passing the additional cost burden on to workers in higher prices.
In Figure 4.24, the price-setting curve shifts down following the oil shock. In this example, a bargaining gap of 2% opens up between the wage-setting curve and the post-shock price-setting curve. This replicates the scenario in Figure 4.20c, where a bargaining gap of 2% appears. This increases inflation from its pre-existing level of 3% to 5% and, as expected inflation adjusts, inflation rises thereafter every year. The Phillips curve shifts up year by year.
Figure 4.24 An oil shock and inflation: the PS curve and Phillips curves.
As long as employment remains at its pre-oil-shock level, inflation will increase every period. The model helps us to understand why a rise in the world price of such an important input to economic activity leads to rising inflation. With a higher oil price, the foreign oil producers claim a larger share of output per worker. Firms at home maintain their share by passing on the cost increase, and workers end up with a smaller share: this is shown by the lower real wage on the new PS curve.
Unless the supply shock is reversed, stable inflation can be restored if the economy operates at the higher structural unemployment rate where the wage on the WS curve is lower and equal to the wage on the post-shock PS curve (shown in Figure 4.20b).
The model also helps to explain the role that high unemployment played in bringing inflation down after the two 1970s oil shocks.
In the model, unless the supply shock is reversed, high inflation can be brought down by a negative bargaining gap. Remember that for the bargaining gap to be negative, unemployment has to rise above the new higher structural unemployment rate. The Phillips curve equation is:
where, for inflation to fall, inflation this period is lower than inflation last period, that is, πt<πt−1, which implies that gapt<0.
In the model, the economy would move along a Phillips curve to the bottom left with inflation falling as unemployment rises. Once inflation has fallen, this will feed into inflation expectations and the Phillips curve will shift downward as inflation expectations are updated.
To understand better how disinflation from an oil shock can occur, we need to bring in the policy maker, which we do in Unit 5.
Exercise 4.11 An oil shock that did not produce rising inflation
There’s a puzzle here: why did the third oil shock from 2002–2008 not lead to increased inflation, just like the earlier ones? This section should have provided you with some starting points to investigate this, and a speech given in 2006 by David Walton, an economist, will help you.2 If you read both carefully, you might ask the following questions:
David Walton. 2006. ‘Has Oil Lost the Capacity to Shock?’. Oxonomics 1 (1): pp. 9–12. Close footnote
- Was the unit cost increase smaller due to less energy-intensive production? This would have meant a smaller increase in the cost of materials per unit of output and reduced the size of the initial downward shift in the price-setting curve.
- Did the wage-setting curve shift downwards at around the same time as the third oil price shock? This also would have reduced or perhaps even eliminated the bargaining gap opened up by the oil price shock. What might have caused the downward shift of the WS curve?
- Did a wage–price spiral fail to develop because expected inflation did not adjust upward, as in the previous oil shocks?
Think about the three questions related to oil shocks that we listed above. In each case:
- Explain the mechanism linking the oil shock to inflation using a diagram.
- Identify some evidence (for example, data or commentary in the economics press) that is consistent with the hypothesis proposed.
-
David Walton. 2006. ‘Has Oil Lost the Capacity to Shock?’. Oxonomics 1 (1): pp. 9–12.