The sky’s the limit: The economics of inflation and hyperinflation

2 Inflation as conflict: My cost is your price, and your cost is my price

To learn more about the circular flow of income, read Section 3.2 of The Economy 2.0: Macroeconomics.

The circular flow of income and expenditure in the economy is a fundamental element of macroeconomics: income = expenditure = the added value of production. The Nobel Prize-winning economist Paul Krugman once captured this idea with the phrase, ‘My spending is your income, and your spending is my income.’¹

This is true at the macro level because it is also true at the level of every individual transaction: whenever a consumer buys a product from a firm, the consumer’s spending is the firm’s income. And whenever a firm pays a worker their wages, the firm’s spending is the worker’s income. This becomes a circular process when the worker spends their wages on firms’ products.

So we can construct the following counterpart to Krugman’s phrase: ‘My cost is your price, and your cost is my price.’

This is the foundation of the idea that price-setting is based on the conflicting interests of different groups in the economy. Since economic actors like to maximize their real incomes, firms like lower wages and higher prices, but workers like lower prices and higher wages. In later sections, we will explore additional conflicts that lead to rising prices. We model changes in inflation as the outcome of a disequilibrium between wages and prices because everyone wants to raise their own price relative to their own costs, which means raising everyone else’s costs relative to their prices.

Modelling inflation using the wage-setting/price-setting model

We start by revisiting the wage-setting/price-setting (WS–PS) model of The Economy 2.0: Macroeconomics (Units 1 and 4), assuming that labour is the only input to production. The wage-setting (WS) curve and price-setting (PS) curve are plotted on a figure showing employment on the horizontal axis and the real wage on the vertical axis (top panel of Figure 4). They show how the total output per worker (productivity, \(\lambda\)) is divided between workers (in the form of wages) and firms (in the form of profits).

To learn more about the wage-setting curve, read Section 1.6 of The Economy 2.0: Macroeconomics.

wage-setting curve, WS curve

A relationship showing, for each level of economy-wide employment, the wage that employers need to set to recruit and motivate workers.

The wage-setting (WS) curve shows the lowest wage that will motivate workers to put in effort, given the economy’s level of employment. The WS curve slopes upwards because firms must pay workers higher real wages to motivate them when unemployment is low: the threat of being fired is less effective when it is easier to find another job. For this reason, workers can bargain for higher wages (or equivalently, employers have to offer higher wages) when unemployment is low, which is also when aggregate demand is high.

To learn more about the wage-setting curve, read Section 1.7 of The Economy 2.0: Macroeconomics.

To understand how capacity constraints affect the PS curve, read Section 4.10 of The Economy 2.0: Macroeconomics.

price-setting curve, PS curve

A relationship showing, for each level of economy-wide employment, the real wage that results when firms maximize profits by setting prices as a constant markup on costs.

price markup

The price minus the marginal cost divided by the price. In other words, the profit margin as a proportion of the price. If the firm sets the price to maximize its profits, the markup is inversely proportional to the elasticity of demand for the good at that price.

The price-setting (PS) curve shows the real wage that maximizes firms’ profits. It is based on the markup that firms use to set their prices and is a horizontal line because we assume that firms set prices as a constant markup on costs. That markup is determined by the degree of competition in the markets for goods and services due to market structure and regulation. Less competition means higher markups, higher profits, and lower real wages. The markup might also depend on the degree of aggregate demand, because when demand is high, firms tend to be capacity constrained. In these situations, firms might behave less competitively: since they are already producing at close to full capacity, they are less likely to attempt to undercut competitors. If this is the case then the PS curve might even slope downwards. The PS curve also depends on competition between firms for workers in the labour market: more competition for workers raises the wage. While the PS curve is so named because it is derived from price-setting behaviour, what it tells us is the level of the real wage: it is the residual output left over for workers once we have accounted for firms’ profits. So it tells us the real wage consistent with firms’ pricing behaviour.

supply-side equilibrium

The economy is in supply-side equilibrium when the markets involved in the production of output are in equilibrium. In the WS–PS model it is the labour market equilibrium; that is, where the price-setting real wage equals the wage-setting real wage.

inflation-stabilizing unemployment rate

The unemployment rate (at supply-side equilibrium) at which inflation is constant. Originally known as the ‘natural rate’ of unemployment. Also known as: structural unemployment rate, non-accelerating rate of unemployment (NAIRU). See also: structural unemployment.

bargaining power

The extent of a person or firm’s advantage in securing a larger share of the economic rents made possible by an interaction.

The WS and PS curves cross when the real wage that workers bargain for is consistent with firms’ markups and implied profits. This is what we call supply-side equilibrium. For reasons we will now discuss, we will usually refer to the unemployment rate at supply-side equilibrium as the inflation-stabilizing unemployment rate \(u^*\).

What happens when unemployment is different from this rate? Suppose that unemployment is below equilibrium, like point B in Figure 4. It produces what we call a positive bargaining gap, meaning that workers will bargain for a higher wage than is consistent with firms’ price-setting behaviour.

To understand the consequences, we can conveniently break down the price determination process into a negotiation between firms’ human resources (HR) departments and their workers, followed by price-setting performed by firms’ marketing departments. First, HR acknowledges that workers need to be paid more so they negotiate a higher nominal wage. But then the marketing department notes that input costs have gone up, so they will raise their price. Since all firms do this, the price level has risen, and the real wage is back to where it was before unemployment fell, on the PS curve.

wage–price spiral

This occurs if an initial increase in wages in the economy is followed by an increase in the price level, which is followed by an increase in wages and so on. It can also begin with an initial increase in the price level.

Both wages and prices have risen, resulting in inflation, and real wages have stayed constant. What happens next? If unemployment remains constant, below the inflation-stabilizing rate, then at the next wage bargaining round workers will again bargain with HR for a higher wage; marketing departments will again put up prices; and we will have further inflation, still with no real wage rise. This is what we call a wage–price spiral.

To learn more about the Phillips curve, read Section 4.5 of The Economy 2.0: Macroeconomics.

The bottom panel of Figure 4 shows the relationship between inflation and unemployment or output, which is known as the Phillips curve. To summarize, higher output and lower unemployment lead to a higher bargaining gap, which leads to higher inflation. When the unemployment level is equal to \(U^*\), the inflation-stabilizing unemployment rate, \(u^* = \frac{U^*}{U^* \, + \, N^*}\), then inflation is stable. If unemployment is above this level, to the left of \(U^*\) (like point C in Figure 4), then inflation is declining. When unemployment is below this level, to the right of \(U^*\), then inflation is rising.

We can safely assume that workers understand this mechanism. So why don’t they demand that firms don’t raise prices, as part of their wage bargain? They cannot, because their real wages depend primarily on prices at other firms. And the workers in a given firm have no incentive to demand their own firm keep its price low, since that price has at most a tiny impact on their own spending power. Indeed, the firm’s workers might be perfectly happy with their own firm raising its price if they see that as necessary for their own wage increase. But since every firm is in the same position, all firms raise their prices, and the result is a rise in the general price level (\(P\)). This is what reduces the real wage \((w = W/P)\) back to its previous level.

The wage–price spiral does not happen all at once. Different firms will often negotiate their wages at different times. This implies that a rise in nominal wages in one firm might imply a temporary rise in real wages for its workers, as long as other firms have not yet gone through the same process. Indeed, if we assume that the timing of different firms’ wage- and price-setting is distributed evenly over time, then the real wage for each worker will take on the sawtooth pattern shown in Figure 5. This is because, over time, an increasing share of firms will have raised their own prices. With individual prices rising at different times, the overall price level rises gradually while each worker’s wage rises in discrete steps. The shaded areas represent the loss of real income, relative to a situation in which prices remained constant at the initial level.

The frequency of wage bargaining

Suppose you are in a wage–price spiral with \(u < u^*\) such that there is a bargaining gap of 2%, meaning that workers will bargain for a 2% real wage rise. As we explained, firms respond by putting up prices by 2%. But this happens at every wage-setting round. So if they bargain once per year, it implies inflation of 2%. If they bargain twice per year, it rises by 4%, and so on. More frequent wage-setting rounds imply higher inflation.

indexation

When wages in contracts rise automatically with the price index.

When economic actors become accustomed to fairly high inflation, they often start to use indexation. This means that wages in contracts rise automatically with the price index. That might occur every year if inflation is relatively low, but the higher inflation becomes, the more frequently workers will want to adjust. As shown in Figure 5, assuming inflation is fixed, more frequent wage bargaining will imply a smaller loss of real income over time. That makes it attractive. However, as we have just argued, the problem is that inflation may depend on the frequency of wage uprating. For example, Brazil passed a law in 1964 indexing both wages and domestic financial assets to inflation on a yearly basis. In 1979, in response to several years of rising inflation, it raised the frequency to every six months. As the model predicts, the result was an even faster increase in inflation.^2 3

Inflation expectations

To learn more about inflation expectations, read Section 4.6 of The Economy 2.0: Macroeconomics.

expected inflation

The belief formed by wage-setters and price-setters about the level of inflation in the next period. See also: inflation.

In our analysis so far, we have used \(u^*\) to denote the inflation-stabilizing rate of unemployment, which is the level of unemployment at which the WS and PS curves cross (also known as the supply-side equilibrium). We do not call it the zero-inflation rate because we assume that there is a baseline level of inflation due to expected inflation: if people expect inflation to be 4%, then contracts and wage bargains will take this into account. So in the model, inflation is equal to the bargaining gap plus expected inflation:

\[\pi_t = \pi_t^E + \text{gap}_t\]

Now suppose that inflation expectations are not fixed but instead are backward-looking or ‘adaptive’, meaning that higher inflation this year leads to higher expected inflation next year:

\[\pi_t^E = \pi_{t-1}\]

In this case, a wage–price spiral leads not just to higher inflation but to ever-rising inflation. If unemployment remains constant at a level below the inflation-stabilizing level, then the inflation rate rises by a gap of \(t\) each year. Prices don’t just rise, they accelerate.

In the Phillips curve diagram, a rise in inflation in year \(t - 1\) raises the Phillips curve for year \(t\), and it will not shift back downwards unless inflation falls. If the bargaining gap remains positive, the Phillips curve continues to shift up and up over time.

inflation target, inflation targeting

Inflation targeting is a form of monetary policy, where the central bank changes interest rates in order to influence aggregate demand and keep the economy close to an inflation target rate, which is normally specified by the government.

On the other hand, we say that expectations are ‘anchored’ if they are equal to some quantity that is, at least in the short term, independent of inflation outcomes. The purpose of an inflation target at some level, \(\pi^T\), particularly when it is implemented by an independent central bank, is to anchor expectations to a fixed figure. When this is successful, we have

\[π_t^E = π^*\]

To learn more about FlexIT and other macroeconomic policy regimes, read Section 7.5 of The Economy 2.0: Macroeconomics.

flexible exchange rate

A country’s exchange rate is flexible if it can change in response to trading in the foreign exchange markets, rather than being held constant by the government or central bank. See also: exchange rate, fixed exchange rate.

credibility

Credibility for a central bank refers to the degree to which economic actors believe that the central bank will do what it says—particularly with respect to commitments to an inflation target or a fixed exchange rate.

When inflation is anchored to the inflation target, a rise in inflation in year \(t - 1\) does not affect the Phillips curve in year \(t\). Many countries have found that an independent central bank with a credible inflation target is an effective way to control inflation, and that it helps to keep expectations anchored to the target. This is the macroeconomic policy regime described as ‘FlexIT’ in Unit 7 of The Economy 2.0: Macroeconomics, where an inflation target is combined with a floating (flexible) exchange rate. But other countries have struggled to foster the credibility that this requires: if people simply do not believe the announced target, then it will not work, and expectations may revert to being adaptive.

fixed exchange rate

A country’s exchange rate is fixed if it is managed by the central bank or the government, and is either held constant over time or kept within a narrow range of values. A country in a common currency zone effectively has a permanently fixed exchange rate relative to all other countries in the zone. See also: exchange rate, flexible exchange rate.

tradable goods

Tradeable goods are those that can be exported, or are close substitutes for imported goods.

adaptive expectations

People have adaptive expectations for inflation when they expect inflation the coming year to be equal to its level last year.

Another way to anchor inflation expectations is through a fixed exchange rate, which makes a foreign currency the price anchor. If domestic currency is pegged to the US dollar (USD), for instance, then the inflation rate of tradable goods should be close to the inflation rate of the USD. This is because these goods compete with imports. Thus, if the price of domestic goods rises faster, then people will choose to buy cheaper imports instead of the more expensive domestic version, forcing domestic producers to keep down their prices. With the inflation rate of tradables anchored to international inflation, the hope is that other prices will follow suit. Implicitly, then, the domestic inflation target is equal to the inflation rate of the currency to which the domestic currency is pegged.

Inflation targets and a fixed exchange rate are unlikely to work as price anchors when there is widespread indexation in the economy. They undermine the possibility of an inflation target because they justify adaptive expectations: if economic actors know that many prices are indexed, then baseline expectations for inflation this year will be close to actual inflation last year. The indexing itself becomes the anchor, crowding out any other proposed anchor.

Reducing inflation

The model above implies that as long as inflation expectations are anchored to a target, then maintaining aggregate demand at the supply-side equilibrium with the unemployment rate at \(u^*\) will always keep inflation on target. If expectations are not well anchored and high inflation is persistent, then in theory the remedy is simple: tightening either fiscal or monetary policy to reduce aggregate demand or ‘cool down’ the economy and raise unemployment, creating a negative bargaining gap, and lowering inflation over time. With luck, expectations will also decline, and once expectations are on target, aggregate demand can be returned to equilibrium with the unemployment rate at \(u^*\) once again.

When inflation is slightly or moderately above target, this involves a slight or moderate recession. Painful for those who lose their jobs, but hopefully temporary, and necessary to avoid more drastic adjustments in the future.

As we will learn in this Insight, however, at high levels of inflation other mechanisms kick in. Then reducing inflation becomes far more difficult and can be far more costly.