The Second Machine Age: Work, Progress, and Prosperity in a Time of Brilliant Technologies (18 page)

Along with wealth, the income distribution has also shifted. The top 1 percent increased their earnings by 278 percent between 1979 and 2007, compared to an increase of just 35 percent for those in the middle of the income distribution. The top 1 percent earned over 65 percent of income in the United States between 2002 and 2007. According to Forbes, the collective net worth of the wealthiest four hundred Americans reached a record two trillion dollars in 2013, more than doubling since 2003.
¹³

I
N
SHORT
, median income has increased very little since 1979, and it has actually fallen since 1999. But that’s not because growth of overall income or productivity in America has stagnated; as we saw in chapter 7, GDP and productivity have been on impressive trajectories. Instead, the trend reflects a significant reallocation of who is capturing the benefits of this growth, and who isn’t.

This is perhaps easiest to see if one compares
average
income with
median
income. Normally, changes in the average income (total income divided by the total number of people) are not very different from changes in median income (income of the person exactly in the middle of the income distribution—half earn more and half earn less). However, in recent years, the trends have diverged significantly, as shown in figure 9.1.

How is this possible? Consider a simple example. Ten bank tellers are drinking beers at a bar. Each of them makes $30,000 a year, so both the mean and median income of this group is $30,000. In walks the CEO and orders a beer. Now the average income of the group has skyrocketed, but the median hasn’t changed at all. In general, the more skewed the incomes, the more the mean tends to diverge from the median. This is what has happened not only in our hypothetical bar but also in America as a whole.

Overall, between 1973 and 2011, the median hourly wage barely changed, growing by just 0.1 percent per year. In contrast, as discussed in chapter 7, productivity grew at an average of 1.56 percent per year during this period, accelerating a bit to 1.88 percent per year from 2000 to 2011. Most of the growth in productivity directly translated into comparable growth in average income. The reason why median income growth was so much lower was primarily because of increases in inequality.
¹⁴

The Three Pairs of Winners and Losers

In the past couple of decades, we’ve seen changes in tax policy, greater overseas competition, ongoing government waste, and Wall Street shenanigans. But when we look at the data and research, we conclude that none of these are the primary driver of growing inequality. Instead, the main driver is exponential, digital, and combinatorial change in the technology that undergirds our economic system. This conclusion is bolstered by the fact that similar trends are apparent in most advanced countries. For instance, in Sweden, Finland, and Germany, income inequality has actually grown more quickly over the past twenty to thirty years than in the United States.
¹⁵
Because these countries started with less inequality in their income distributions, they continued to be less unequal than the United States, but the underlying trend is similar worldwide across sometimes markedly different institutions, government policies, and cultures.

As we discussed in our earlier book
Race Against the Machine
, these structural economic changes have created three overlapping pairs of winners and losers. As a result, not everyone’s share of the economic pie is growing. The first two sets of winners are those who have accumulated significant quantities of the right capital assets. These can be either nonhuman capital (such as equipment, structures, intellectual property, or financial assets), or human capital (such as training, education, experience, and skills). Like other forms of capital, human capital is an asset that can generate a stream of income. A well-trained plumber can earn more each year than an unskilled worker, even if they both work the same number of hours. The third group of winners is made up of the superstars among us who have special talents—or luck.

In each group, digital technologies tend to increase the economic payoff to winners while others become less essential, and hence less well rewarded. The overall gains to the winners have been larger than total losses for everyone else. That simply reflects the fact we discussed earlier: productivity and total income have grown in the overall economy. This good news offers little consolation to those who are falling behind. In some cases the gains, however large, have been concentrated among a relatively small group of winners, leaving the majority of people worse off than before.

Skill-Biased Technical Change

The most basic model economists use to explain technology’s impact treats it as a simple multiplier on everything else, increasing overall productivity evenly for everyone.
¹⁶
This model can be described in mathematical equations. It is used in most introductory economics classes and provides the foundation for the common—and until recently, very sensible—intuition that a rising tide of technical progress will lift all boats, that it will make all workers more productive and hence more valuable. With technology as a multiplier, an economy is able to produce more output each year with the same inputs, including labor. And in the basic model all labor is affected equally by technology, meaning every hour worked produces more value than it used to.

A slightly more complex model allows for the possibility that technology may not affect all inputs equally, but rather may be ‘biased’ toward some and against others. In particular, in recent years, technologies like payroll processing software, factory automation, computer-controlled machines, automated inventory control, and word processing have been deployed for routine work,
substituting
for workers in clerical tasks, on the factory floor, and doing rote information processing.

By contrast, technologies like big data and analytics, high-speed communications, and rapid prototyping have
augmented
the contributions made by more abstract and data-driven reasoning, and in turn have increased the value of people with the right engineering, creative, or design skills. The net effect has been to decrease demand for less skilled labor while increasing the demand for skilled labor. Economists including David Autor, Lawrence Katz and Alan Krueger, Frank Levy and Richard Murnane, Daron Acemoglu, and many others have documented this trend in dozens of careful studies.
¹⁷
They call it
skill-biased technical change
. By definition, skill-biased technical change favors people with more human capital.

The effects of skill-biased technical change can be vividly seen in figure 9.2, which is based on data from a paper by MIT economists Daron Acemoglu and David Autor.
¹⁸
The lines tell a story about the diverging paths of millions of workers over recent generations. Before 1973, American workers all enjoyed brisk wage growth. The rising tide of productivity increased everyone’s incomes, regardless of their educational levels. Then came the massive oil shock and recession of the 1970s, which reversed the gains for all groups. However, after that, we began to see a growing spread of incomes. By the early 1980s, those with college degrees started to see their wages growing again. Workers with graduate degrees did particularly well. Meanwhile, workers without college degrees were confronted with a much less attractive labor market. Their wages stagnated or, if they were high school dropouts, actually fell. It’s not a coincidence that the personal computer revolution started in the early 1980s; the PC was actually
Time
magazine’s “machine of the year” in 1982.

The economics of the story become even more striking when one considers that the number of college graduates grew very rapidly during this period. The number of people enrolled in college more than doubled between 1960 and 1980, from 758,000 to 1,589,000.
¹⁹
In other words, there was a large increase in the supply of educated labor. Normally, greater supply leads to lower prices. In this case, the flood of graduates from college and graduate school should have pushed down their relative wages, but it didn’t.

The combination of higher pay despite growing supply can only mean that the relative
demand
for skilled labor increased even faster than supply. And at the same time, the demand for tasks that could be completed by high school dropouts fell so rapidly that there was a glut of this type of worker, even though their ranks were thinning. The lack of demand for unskilled workers meant ever-lower wages for those who continued to compete for low-skill jobs. And because most of the people with the least education already had the lowest wages, this change increased overall income inequality.

Organizational Coinvention

While a one-for-one substitution of machines for people sometimes occurs, a broader reorganization in business culture may have been an even more important path for skill-biased change. Work that Erik did with Stanford’s Tim Bresnahan, Wharton’s Lorin Hitt, and MIT’s Shinkyu Yang found that companies used digital technologies to reorganize decision-making authority, incentives systems, information flows, hiring systems, and other aspects of their management and organizational processes.
²⁰
This coinvention of organization and technology not only significantly increased productivity but tended to require more educated workers and reduce demand for less-skilled workers. This reorganization of production affected those who worked directly with computers as well as workers who, at first glance, seemed to be far from the technology. For instance, a designer with a knack for style might find herself in greater demand at a company with flexible equipment in distant factories that can quickly adapt to the latest fashions, while an airport ticket agent might find himself replaced by an Internet website he never knew existed, let alone worked with.

Among the industries in the study, each dollar of computer capital was often the catalyst for more than ten dollars of complementary investments in “organizational capital,” or investments in training, hiring, and business process redesign.
²¹
The reorganization often eliminates a lot of routine work, such as repetitive order entry, leaving behind a residual set of tasks that require relatively more judgment, skills, and training.

Companies with the biggest IT investments typically made the biggest organizational changes, usually with a lag of five to seven years before seeing the full performance benefits.
²²
These companies had the biggest increase in the demand for skilled work relative to unskilled work.
²³
The lags reflected the time that it takes for managers and workers to figure out new ways to use the technology. As we saw in our earlier discussion of electrification and factory design, businesses rarely get significant performance gains from simply “paving the cowpaths” as opposed to rethinking how the business can be redesigned to take advantage of new technologies.
²⁴
Creativity and organizational redesign are crucial to investments in digital technologies.
*

This means that the best way to use new technologies is usually not to make a literal substitution of a machine for each human worker, but to restructure the process. Nonetheless, some workers (usually the less skilled ones) are still eliminated from the production process and others are augmented (usually those with more education and training), with predictable effects on the wage structure. Compared to simply automating existing tasks, this kind of organizational coinvention requires more creativity on the part of entrepreneurs, managers, and workers, and for that reason it tends to take time to implement the changes after the initial invention and introduction of new technologies. But once the changes are in place, they generate the lion’s share of productivity improvements.

The Skill Set Affected by Computerization Is Evolving