Chapter 10 The interaction of the supply and the demand side: economic shocks and inflation

10.1 Goods market and distributional equilibrium

We can now combine the demand side developed in Part II with the supply side of our basic model presented in Part III and analyse interactions between the demand and supply sides. Here, we assume a given inflation-stable employment (\(L^N\)) or NAIRU determined by the given institutional factors and labour and goods market norms, as we derived in chapter 9. We also exclude for the moment the coordination of wage bargaining as a way to raise \(L^N\) and reduce the NAIRU. Economic policy interventions for stabilisation are also not considered for now, since we first want to understand what macroeconomic outcomes the interaction of market processes in the labour and goods markets leads to in our model.

In the figure 10.1, the demand side of the overall model is on the left and the supply side of the overall model is on the right. In the following description, we move clockwise through the quadrants of figure 10.1:

  • We start at the bottom with the \(IS\) curve in the lower quadrant on the left-hand side. For a given exogenous interest rate, GDP is determined in the goods market equilibrium.
  • In the upper quadrant on the left-hand side, the same situation is shown for the income-expenditure model: At the intersection of the aggregate demand curve and the 45-degree line, the goods market is in equilibrium.
  • Since at equilibrium the supply of goods has adjusted to the demand for goods, the production function in the upper quadrant on the right-hand side makes it possible to determine the level of employment belonging to the goods market equilibrium.
  • We have chosen the values here so that the employment level of the goods market equilibrium, \(L^*\), coincides with the employment level of the distribution equilibrium, \(L^N\), which is determined in the middle quadrant on the right-hand side by the intersection of the \(WS\) curve with the \(PS\) curve. In this quadrant, the target real wage rates of workers and firms are represented in the \(WS\) and \(PS\) curves. The intersection of these two curves determines the employment level of the distributional equilibrium.
  • In the lower quadrant on the right-hand side, we see the short-run Phillips curve that results from the \(WS-PS\) quadrant in the middle. Here we can now read which inflation rate results depending on the employment level, which is determined in the goods market.

Figure 10.1: Demand and supply side of the complete model.

We can now also use this representation to analyse the effects of a change in the goods market equilibrium. In figure 10.2, for a given interest rate, there is a positive demand shock, i.e. a rightward shift of the \(IS\) curve in the lower quadrant on the left-hand side or a shift of the aggregate demand curve in the upper quadrant on the same side. Higher GDP in the goods market equilibrium is now associated with higher employment, as illustrated by the production function in the upper quadrant on the right-hand side of figure 10.2. However, higher employment now leads to the real wage target of workers being higher than that of firms, as illustrated by the middle quadrant on the right-hand side. This now causes the inflation rate to rise in line with the Phillips curve in the lower quadrant on the right-hand side compared with the initial situation. Rising (or falling) inflation is therefore an expression of the fact that the employment level determined by the goods market equilibrium deviates from the employment level of the distributional equilibrium. If such a deviation exists, however, the process does not end with a higher (lower) inflation rate, as we will show in the next section.

Figure 10.2: Demand and supply side of the complete model after a positive demand shock.

10.2 Demand shocks and wage-price spirals

In the previous section, we learned about the basic mechanisms by which a change in demand and employment lead to a change in the inflation rate. Here we discuss what happens when the goods market equilibrium level of employment deviates from that of the distributional equilibrium, the inflation rate changes as a result and there is no immediate reaction of economic policy.

Let us again assume a positive demand shock. The increase in demand leads to an expansion of output, and firms hire more workers. The increased employment increases the bargaining power of workers and they are able to push through a higher nominal wage increase in the current wage round. For their part, companies respond by raising the inflation rate, thereby enforcing their real wage (or unit profit) expectations. The positive demand shock thus leads to a higher inflation rate - up to this point, the process is identical to the one described in the previous section. But what happens now? Here we assume in the following that the real interest rate does not change as a result of the increase in inflation, and that the nominal interest rate is therefore adjusted immediately in line with the increase in the inflation rate.

We still have a situation in the goods market and in the labour market in which employment is above the level of the distribution equilibrium. Although workers have not been able to push through their real wage demands because companies have counteracted them with price increases, the bargaining power of workers is still at the same level. So they will again try to raise the real wage to the desired level through corresponding nominal wage increases. But from the perspective of workers, how much does the nominal wage have to increase in the current round in order to reach the desired real wage level?

As we saw in the derivation of the Phillips curve, this depends, among other things, on the expected inflation rate on which workers base their nominal wage demands. However, since workers have already observed an increase in the inflation rate in the last period, they will adjust (adapt) their inflation expectations accordingly in the current round of negotiations. In our simple model of expectation formation (adaptive expectations), they will assume that inflation in the current period will assume exactly the value of the previous period (\(\pi^e = \pi_{-1}\)). Thus, employees assume a higher level of inflation for their nominal wage demands than in the first wage round, which leads to nominal wages rising at a higher rate than before. Once again, companies will now react in the same way as in the first round and raise prices in line with the now higher growth rate of nominal wages.

The inflation rate rises again, although we have not observed another demand shock. Instead of a change in demand or employment, the renewed surge in inflation is now based solely on an increase in inflation expectations and the still unresolved distribution conflict. In the first round, we had moved along the short-run Phillips curve because of the rise in employment: in the current round, the rise in inflation expectations instead leads to an upward shift of the short-run Phillips curve.36 It is now also clear why we have called these short-run Phillips curves. Each curve applies only to a particular set of inflation expectations. As inflation expectations change from period to period outside the distributional equilibrium, the short-run Phillips curves shift. Employment remains constant. The figure 10.3 depicts this shift in the Phillips curve, caused by the adjustment of inflation expectations, in the second round after the increase in demand.37

Figure 10.3: Demand shock and inflation.

The gap between the old and the new short-run Phillips curve is given by the change in inflation expectations (in our example, this is identical to the change in inflation itself). Since the employment situation in the labour market has still not changed, the renewed surge in inflation will lead to a further increase in inflation expectations in the next round. The Phillips curve would accordingly shift upward again, and this process would be repeated in each future round. The Phillips curve would continue to shift upward, and the inflation rate would rise at an ever-increasing rate. This is why we refer to it as accelerating inflation. Figure 10.4 shows such a process over several rounds.

Figure 10.4: Positive demand shock and rising inflation.

Without a change in the behaviour described above, an initial expansionary demand shock can thus lead to a repeated rise in the inflation rate. The central cause here is the unresolved distributional conflict between wages and profits. Since this conflict is played out through nominal wages and prices, we also refer to this phenomenon as a wage-price spiral. As we will see, this process can be terminated by rolling back demand in the goods market and employment in the labour market to their original levels.

Suppose, for example, that demand falls back to its original level after some time (e.g., in the second round) - we then speak of a temporary demand shock - so that a distributional equilibrium is re-established. Employment thus falls back to its inflation-stabilising level, which is still given by the intersection of the wage-setting and price-setting curves. Figure 10.5 shows this situation in the \(WS-PS\) and Phillips curve diagram.

Figure 10.5: Temporary demand shock and inflation rate.

In the \(WS-PS\) diagram, employment in the second round has returned to the distributional equilibrium level after the temporary shock. Since there is thus no longer any conflict between the real wage expectations of companies and the unions or workers, the rate of change in nominal wages and the inflation rate stabilise at a constant level. However, since the inflation expectations of workers have shifted upward - visible in the figure from the shift in the short-run Phillips curve - the level of inflation is significantly higher than in the initial situation.

In the event of a negative demand shock, i.e. a drop in demand that may be triggered by rising interest rates or a deterioration in companies’ sales expectations, for example, the mechanisms described above act in reverse. The drop in demand leads to a drop in employment and thus to a deterioration in the bargaining position of workers. Employees are now prepared to accept lower real wages and reduce their nominal wage demands accordingly. The decline in unit labour cost growth and the competition among companies cause them to adjust the inflation rate of their products to the lower nominal wage inflation. Thus, firms lower the inflation rate so that the target real wage, which they calculate as optimal, returns. The fall in the inflation rate - we also speak of disinflation - now also leads to an adjustment of the inflation expectations of workers. This results in another shift of the Phillips curve, but this time downward. In the figure 10.6, the Phillips curve accordingly shifts successively downward and the inflation rate falls from round to round. In our numerical example shown, the inflation rate finally slips into negative territory in the sixth round. Disinflation turns into deflation.

Figure 10.6: Negative demand shock: disinflation and risk of deflation.

When the average price level falls, we speak of deflation. The inflation rate is then in negative territory. Deflation is a rather rare phenomenon compared to inflation (at least from the 20th century onwards), but deflation can pose a major challenge to economic policy. Deflation often occurs in the context of deep economic crises and it can exacerbate crises via real debt effects and by postponing spending decisions. A crisis accompanied by deflation thus places high demands on economic policy, as explained below.

In the following interactive app, wage-price spirals can be triggered by a positive or negative demand shock.

10.3 Supply shocks

In addition to the demand shocks discussed above, our economy can also be subject to a supply shock. A supply shock can have various causes. While a demand shock in our model acts via the \(IS\) curve or aggregate income and expenditure, a supply shock can act either via the production function or via the labour market in the \(WS-PS\) diagram. In our model, given constant short-run production technology and thus constant labour productivity, there are basically two different types of supply shocks:

  1. A labour market shock leads to a change in the position of the \(WS\) curve, i.e., the target real wage curve of employees. This can be caused either by a change in \(\mathbf{b}\) (e.g., a reduction due to a cut in benefits) or by a change in the conflict orientation, \(k\), of employees (e.g., a reduction due to a legal weakening of unions).
  2. a price-setting shock that leads to a shift in the \(PS\) curve, i.e., the target real wage rate curve of firms. This is caused by a change in the mark-up, e.g., a decrease due to an increase in the intensity of competition in the goods market.38

Let us first start with a shift in the wage-setting curve due to a cut in social benefits. The minimum required real wage of employees, \(\mathbf{b}\), falls and the wage-setting curve shifts downward as a result, as illustrated in figure 10.7:

Figure 10.7: Positive supply shock.

The downward movement of the wage-setting curve shifts the distributional equilibrium in figure 10.7 to the right. Thus, inflation-stabilising employment, \(L^N\), is now higher than before and is now higher than actual employment, \(L^*\), which was determined by the goods market. The gap created between the wage-setting curve and the price-setting curve now triggers a disinflation process. At the current level of employment, the target real wage rate of employees is now below the target real wage rate of firms. The resulting decline in nominal wage growth is therefore passed on by firms through prices. The inflation rate thus falls, although unlike a demand shock, we do not observe a movement along the Phillips curve here, but a downward shift of the Phillips curve, from \(PC_0\) to \(PC_1\), which already takes place in the first round. In the next round, the decline in the inflation rate also changes the inflation expectations of employees, and the further course of events resembles the process caused by a negative demand shock. As inflation expectations fall, the Phillips curve shifts further and further downward, threatening a process of ever more sharply falling inflation. Without intervention, there is also the threat of a disinflationary spiral, which in extreme cases can lead to deflation.

Thus, although the supply shock has negative effects on the economy in the absence of policy countermeasures, we usually refer to this as a positive supply shock. The positive aspect of this shock is its effect on the distributional equilibrium and thus on inflation-stabilising employment, \(L^N\). Indeed, stable inflation rates can now be achieved at a higher level of employment, or a lower unemployment rate. Thus, a positive supply shock increases inflation-stabilising employment and, hence, potential output and employment. However, there is no guarantee that the economy will actually reach this higher potential employment level. Without economic policy intervention, the positive supply shock is more likely to lead to disinflation and deflation, with then possibly negative repercussions on effective demand in the goods market and employment.

But is a cut in social benefits really purely a supply shock? A cut in social benefits is often portrayed (as above) as a pure supply shock. However, this portrayal is to some extent misleading because it ignores a possible effect on demand. However, since social benefits are to a large extent disposable income of (poorest) households, their consumption expenditure will be reduced by the cuts. The same would apply, for example, to a cut in the minimum wage, which would also shift the wage-setting curve downward. As a direct consequence, the resulting demand shortfall leads to increased unemployment and further amplifies the disinflationary push triggered by the shift in the wage-setting curve.

The economy can, of course, also be subject to a negative supply shock. A negative supply shock occurs, for example, when the wage-setting curve shifts upward. Reasons for this can be, for example, an increase in social benefits or the minimum wage. However, there are also other factors that influence the minimum real wage demanded (for example, other institutional factors or social norms and values). A negative supply shock shifts the distributional equilibrium to the left; inflation-stabilising employment is now lower than before. At current employment levels, employees demand a higher real wage, and the associated higher nominal wage increases lead to a process of increasingly rising inflation, largely analogous to a positive demand shock.

Figure 10.8: Negative supply shock.

A price-setting shock, on the other hand, can occur as a result of a change in the mark-up, \(m\), of firms. This can be triggered, for example, by an increase in competition in the goods market, e.g. by opening markets to additional competitors. This leads to an upward shift in the price-setting curve; firms’ target real wage rate rises, as figure 10.9 shows. This increases inflation-stable employment, and the NAIRU declines. Actual employment is now below inflation-stable employment, and nominal wage growth and the inflation rate decline. The short-run Phillips curve shifts downward and a disinflation process sets in, which can again lead to deflation. The following figure illustrates a shift in the price-setting curve caused by a mark-up shock.

Figure 10.9: Positive price shock.

So far, we have assumed here that the nominal interest rate automatically adjusts to prevailing inflation, so that the real interest rate remains constant. However, if we abandon this assumption and instead assume that the nominal interest rate remains constant, this would also lead to a change in the real interest rate in the event of a change in inflation. This, in turn, would have an effect on investment demand. With a constant nominal interest rate, the rise (fall) in the inflation rate would lower (raise) the real interest rate through a positive (negative) demand shock. The positive (negative) effect on demand would then move the goods market equilibrium further away from the distributional equilibrium. The following interactive app can be used to simulate the supply and demand shocks described above, assuming here the very same assumption of a constant nominal interest rate.

Further reading on chapter 10

Textbooks:

Literature

Carlin, W., and D. W. Soskice. 2015. Macroeconomics: Institutions, Instability, and the Financial System. Oxford University Press.

  1. We show the Phillips curve in the initial situation and the Phillips curve in round 1 overlapping each other. In fact, this would be the same curve in two subsequent rounds, since inflation expectations have not yet changed.↩︎

  2. In the figures of the Phillips curve, the curve of the current round is shown in blue. The Phillips curve in the baseline and the Phillips curve of the previous rounds appear in black.↩︎

  3. A commodity price shock, e.g., an oil price shock, could also be illustrated via a mark-up shock, since the mark-up in our model must cover all production costs that are not part of variable wage costs.↩︎