Unit 2 Unemployment, wages, and inequality: Supply-side policies and institutions
2.9 Application: Did a decline in competition increase inequality in the US?
Governments use competition policies to address microeconomic inefficiency in markets for particular goods, in which firms with market power set a price above marginal cost, which results in too little of a good being produced (read Section 7.12 of the microeconomics volume). As demonstrated in Section 2.3, the macro model reveals a broader harm to the economy of excessive market power. The implication is that policies that increase competition in the product market reduce firm markups and reduce inequality by increasing the wage share (and real wages) as well as by reducing structural unemployment.
In spite of competition policies (also known as antitrust regulation), firms have become more powerful in relation to their customers since the 1980s in the US. And we can use the macro model to show how this phenomenon increases economic inequality. A fall in the degree of competition in markets for goods and services, ceteris paribus, entails higher structural unemployment and a higher share of profits. Both raise inequality among households.
Recent research shows a rise in the markup in the US, and in many other countries. Figure 2.21 for the US shows a falling average markup from the mid-1960s to 1980, a rapid increase until 2000, and then, after a decade of stability, a renewed rise since the global financial crisis. The average markup is more than twice now what it was in 1980.
For an interesting comparison with the data for Europe, read Jan Eeckhout. 2021. ‘Book review: The great reversal by Thomas Philippon’. Journal of Economic Literature 59 (4): pp. 1340–1960.
Figure 2.21 The estimated average markup for firms in the US (1955–2016).
Jan De Loecker, Jan Eeckhout, and Gabriel Unger. 2020. ‘The Rise of Market Power and the Macroeconomic Implications’ The Quarterly Journal of Economics 135 (2): pp. 561–644.
Note: The chart shows our measure of the markup (\(μ\)); the paper uses a different measure, which is equal to \(1/(1-μ)\).
- profit, economic profit
- A firm’s profit is its revenue minus its total costs. We often refer to profit as ‘economic profit’ to emphasise that costs include the opportunity cost of capital (which is not included in ‘accounting profit’).
Over the same period, the share of the economy’s income going as economic profits to the owners of firms has been increasing, as illustrated also for the US in Figure 2.22.
Figure 2.22 The share of economic profits in income in the US non-financial corporate sector (1946–2016).
Simcha Barkai. 2020. ‘Declining Labor and Capital Shares’. The Journal of Finance 75 (5): pp. 2421–2463; S. Barkai & S. G. Benzell. ‘70 years of US corporate profits’. Working Paper.
In estimating the profit share for the US economy in Figure 2.22, the authors divide income into three parts. One is the labour share. The rest is ‘profits’, which are split into the other two shares. The ‘capital’ share is the opportunity cost of capital as a share of income; the remainder is what is labelled in the chart ‘Profit share’ and is the share of economic profits in income. In the WS–PS model, there is no capital, which means that the share of economic profits is the profit share.
Listen to John Van Reenen, an economist, talking about the rise of superstar firms and the challenges they pose for policymakers aiming to sustain or restore a competitive economy.
The research on trends in both markups and the share of economic profits points to the central role of declining competition in markets for goods and services. This development suggests that there has been a long-term shift in the balance of power in the US economy (and in some other countries, including in Europe) towards the owners of firms and away from workers and their customers. At the same time—from the 1980s—the inequality among US households has been on a rising trend. An important question, which is still much debated, is the extent to which falling competition contributed to this trend.
Figure 2.23 shows the changes in inequality among US households in their market income (before the payment of taxes and receipt of transfers) from 1913 to 2019.
Figure 2.23 The Gini coefficient for market income in the US (1913–2019).
World Inequality Database. 2023.
The data on rising markups and profit share in the US since 1980 is consistent with a decline in the extent of competition faced by firms. In the WS–PS model, this is represented by a downward shift in the price-setting curve. Just considering the shift in the PS curve, the new equilibrium is at point B in Figure 2.24. The model predicts that a decline in the degree of competition and the resulting downward shift in the PS curve will bring about a rise in inequality for two reasons:
- A higher markup means a higher profit share: In the model with only labour costs, the markup is the profit share, as shown in the new lower PS curve in Figure 2.24 and the fall in the wage share from 0.79 to 0.60.
- Unemployment in equilibrium is higher: In the figure for an economy of 100 people, the number of unemployed people rises from 7 to 10.
However, this prediction poses a puzzle: the predicted rise in structural unemployment at point B did not occur in the US economy. To reconcile the model with the facts of falling competition in product markets and, if anything, falling unemployment, we need to incorporate another development that took place in the US over the same period, namely the weakening of labour unions. This was documented in Section 2.5. In the WS–PS model, as shown in Figure 2.14, weaker unions mean a downward shift of the wage-setting curve, which on its own would reduce structural unemployment.
Work through the analysis in Figure 2.24 to understand how the combination of growing market power, with a weaker bargaining position for workers, aligns with the model’s prediction that inequality rises, without a rise in unemployment. Indeed, the example shown in the figure (point C) has lower unemployment than the initial situation at point A.
Figure 2.24 Rising market power (the product market) and declining worker power (the labour market)—a new equilibrium.
The initial situation
The new price-setting curve
The new wage-setting curve
Question 2.10 Choose the correct answer(s)
Suppose that compared to a decade ago, the real wage in an economy has increased and unemployment has increased, while the size of the labour force remains unchanged. Which of the following changes in the WS–PS model are consistent with this information?
- The PS curve would shift downwards and the WS curve would shift upwards, so the real wage would fall but unemployment would rise.
- The PS curve would shift upwards and the WS curve would shift downwards, so the real wage would rise but unemployment would fall.
- The PS curve would shift upwards and the WS curve would shift upwards, so real wages would rise and unemployment would increase.
- Both changes only affect the PS curve, so the new equilibrium will either have higher real wages and lower unemployment or lower real wages and higher unemployment.
Using the model developed in Section 2.3, we can now illustrate the implications for inequality of the rise in market power combined with the weakening of trade unions, as well as the emergence of the gig economy (Figure 2.25).
In this example, the impact of the rise in the profit share in pushing the Lorenz curve away from the line of equality outweighs the impact of the fall in unemployment in reducing inequality.
Figure 2.25 Rising market power (the product market) and declining worker power (the labour market) leads to higher inequality in the new equilibrium.
Lorenz curves in the initial and new equilibrium
The decline in the wage share
The fall in unemployment
The combination of the model (Figures 2.24 and 2.25), with the data on markups, and the profit share in Figures 2.21 and 2.22, support the hypothesis that falling competition in markets for goods and services in the US is part of the reason for rising inequality among households, as shown by the Gini coefficient in Figure 2.23. The trends in these charts are similar, but other factors may still be at work.
Other important factors relevant to the rise in the inequality of household market incomes in the US are automation (The Economy 1.0 Section 19.7), and the so-called China shock (The Economy 1.0 Section 18.7). People in some occupations such as clerical work have been the losers from the development of new technologies that replace routine jobs. The China shock occurred when goods manufactured in China came on to the world market after it joined the World Trade Organization in 2001. As a consequence, those working in particular industries like furniture, garments, toys, and electronic goods lost their jobs. The losers from automation and the China shock moved into less secure, lower-paid work or left the labour force.
Exercise 2.8 Automation and inequality
Suppose an economy initially has five employers and 95 workers. Five of the workers are unemployed, and among the 90 who are employed, all receive the same wage, whether they do routine or non-routine work. These 90 workers receive 60% of the economy’s output, and the employers receive the remaining 40%.
- Draw a WS–PS diagram and Lorenz curve diagram to illustrate this situation and calculate the corresponding Gini coefficient.
Now suppose that machines are introduced. These machines perform routine operations that had always been done by humans. As a result, of the workers who do routine work, five become unemployed and 55 suffer a fall in their bargaining power (because they too can be replaced) and now only receive 10% of the economy’s output. 30 workers (who do non-routine work and cannot be replaced by machines) receive the remaining 50%. - Use your Lorenz curve diagram from Question 1 to illustrate this situation and calculate the corresponding Gini coefficient.
