978-0133780581 Chapter 13 Lecture Note

Unlock access to all the studying documents.

View Full Document

Chapter 13. Technological Progress: The Short, the Medium, and the long run

I. MOTIVATING QUESTION

Does technological progress hurt workers?

In the medium run, technological progress does not create unemployment, in theory or in evidence; in the

short run, technological progress sometimes reduces unemployment, sometimes increases it.

Technological progress does affect income distribution. Many economists believe that the skill bias of

technological change has been the primary reason for the sizeable increase in U.S. income inequality over

the past three decades.

How do institutions affect productivity growth?

Institutions that protect property rights create an environment more conducive to productivity growth, but

causality runs in both directions. If better institutions lead to faster growth, faster growth also leads to

better institutions. Poor countries find it difficult to create institutions that support growth.

II. WHY THE ANSWER MATTERS

Chapters 10 through 12 established technological progress as the determinant of long-run growth per

person. Nevertheless, workers often fear that technological progress will eliminate their jobs. In response

to this concern, Chapter 13 argues that productivity growth does not appear systematically to increase the

unemployment rate. However, technological progress may affect the distribution of income.

Most economists believe that developing better institutions is a prerequisite for growth in poor

economies.

III. KEY TOOLS, CONCEPTS, AND ASSUMPTIONS

1. Tools and Concepts

Essentially, there are no new tools in this chapter. The wage-setting, price-setting framework of Chapter 7

is adjusted to include productivity.

2. Assumptions

Since the focus of this chapter is unemployment, the text assumes that labor is the only factor of

production.

IV. SUMMARY OF THE MATERIAL

13-62

1. Productivity, Output, and Unemployment in the Short Run

To focus on employment issues, ignore capital, and write the production function as

Y = AN. (13.1)

Since there is no capital, there is no distinction between technological progress and productivity growth in

this chapter. An increase in A reduces the cost of producing a given level of output (because less labor is

required). The introduction of new technology (A) may lead consumers to spend more, in expectation of

higher future income, and may lead firms to invest more, in anticipation of higher future profits. These

effects would cause IS to shift right. On the other hand, if the increase in A is a result of better use of

existing technology, workers may worry about the safety of their jobs, leading them to save more. In this

case, IS would shift left.

Equation (13.1) implies that the effect of productivity growth on employment is ambiguous: employment

rises (falls) when output grows by a greater (lesser) percentage than productivity and remains constant

when output grows at the same rate as productivity. Even if AD shifts right in response to an increase in

productivity, employment could still fall in the short run.

Empirically, changes in output in the United States tend to be larger than changes in labor productivity.

This fact does not imply that increases in productivity increase employment, because causation also runs

the opposite way—from output growth to productivity growth—as a result of labor hoarding. Since firms

tend to keep extra workers in bad times, measured productivity tends to increase in good times simply

because firms make more use of their employees.

2. Productivity and the Natural Rate of Unemployment

In the medium run, unemployment returns to its natural rate, determined by the wage-setting and

price-setting relations. Productivity growth affects both equations. Given the production function in

equation (13.1), the marginal cost of production is W/A. Thus, the price-setting relation becomes

P = (1+m)W/A. (13.3 on page 267)

The text notes that empirical evidence suggests that wage setting reflects productivity growth. This

observation suggests an extension of the wage-setting equation. Letting (Ae) denote expected productivity

growth, the new wage-setting equation is

W = AePeF(u,z). (13.4)

When price and productivity expectations are correct, equations (13.3) and (13.4) can be written as

equations (13.4) and (13.5) below.

W/P=A/(1+m)

(13.5)

W/P=AF(u,z) (13.6)

The solution to these two equations implies that the natural rate of unemployment is independent of the

level or growth rate of productivity (i.e., A cancels out). In terms of the wage-setting, price-setting

diagram of Chapter 7, an increase in A shifts the wage-setting and price-setting curves up by the same

proportion at the initial natural rate of unemployment, so the natural rate does not change.

13-63

Empirically, a plot of the decade averages of labor productivity growth and the unemployment rate for the

United States over the period 1890-2014 suggests that the two variables are basically unrelated. If

anything, there may be a weak negative relationship (i.e., when productivity growth increases, the

unemployment rate falls). The negative relationship can be reconciled with the preceding analysis if

expectations of productivity growth (which affect wage setting) lag behind actual changes (which affect

price setting). If so, a period of slowing productivity growth would be associated with a higher natural

rate of unemployment. To see this, suppose price expectations are correct, substitute Ae for A in equation

(13.4), start from an original medium-run equilibrium where Ae=A , and assume that A falls, so that Ae>A.

Then, at the original natural rate (where Ae=A), the real wage demanded by wage setters exceeds the real

wage paid by price setters. The natural rate will rise to reduce the real wage demanded by wage setters.

When Ae falls to match the fall in A, the natural rate will return to its original value.

In the late 1990s, high productivity growth was accompanied by optimism about the future, so spending

by firms and consumers was high, output growth exceeded productivity growth by a substantial margin,

and the unemployment rate fell. In the “jobless recovery” of 2002 and 2003, exceptionally high

productivity growth was accompanied by skepticism about the New Economy, so spending was relatively

low, and output growth was not high enough to prevent a rise in the unemployment rate. Finally,

consistent with the earlier results in this chapter, the unexpected increase in productivity growth since the

mid-1990s has been associated with a lower average rate of unemployment, and it seems, with a lower

natural rate of unemployment. The results of the chapter suggest that the effect of productivity growth on

the natural rate will disappear over time. Since high productivity growth is no longer unexpected, the

modified wage-setting equation implies that wages will eventually increase to capture the benefits of

higher worker productivity. Recall, however, that in Chapter 8 we discussed several factors (including

productivity growth) that may have played a role in reducing the U.S. natural rate of unemployment.

Some of these factors may persist. Note that a structural change in the economy that leads to job loss or

lower wages does potentially create negative perceptions about technology and can lead to changes in

worker expectations.

3. Technological Progress, Churning, and Distributional Effects

Even if technological progress helps the economy generally and does not lead to increases in the

aggregate unemployment rate, it still may hurt some individuals. In the United States, for example, wage

inequality has increased substantially over the past 20 years. Workers with a high level of education have

enjoyed an increase in their relative real wage; workers with a low level of education have suffered a

decrease in their relative real wage. Indeed, workers with the least amount of education have actually

suffered a decrease in the absolute level of their real wage. The increase in the relative wage of high-skill

workers reflects increased relative demand for high-skill workers.

Two common explanations for this phenomenon are increased international trade, which exposes

low-skill U.S. workers to foreign competition, and skill-biased technological progress. The trade

explanation, however, does not explain why the relative demand for high-skill workers seems to have

increased even in those sectors not exposed to foreign competition. This observation, among other

reasons, has led most economists to emphasize skill-biased technological progress as the primary

explanation for increasing wage inequality. The process of growth that develops new goods and

simultaneously makes other goods obsolete has been dubbed creative destruction by Joseph Schumpeter.

There is also reason to suspect that the trend will not continue indefinitely. In the first place, the shift in

relative demand for high-skill workers may slow down. Computers may begin to replace high-skill

workers, or firms may be motivated to explore new technologies that make use of relatively inexpensive

low-skill workers. Moreover, the relative supply of high-skill workers may increase, since the wage

differential may motivate more workers to invest in education.

13-64

V. PEDAGOGY

This chapter is optional. It can be omitted without loss of continuity. On the other hand, this chapter is

relatively accessible to students, and can be included in a course even if the chapters on growth are

omitted.

VI. EXTENSIONS

Problems at the end of Chapter 13 ask students to incorporate productivity and expected productivity into

the AS relation. Instructors may wish to discuss this topic in class.

Behavioral economics can also be introduced in this chapter with regard to individual fears regarding

technology.

13-65