First, we need to think about what we mean by "wages." A worker's nominal wages are the dollar value of her take-home pay, not including benefits. For example, if a worker earns $20 per hour before taxes, that is her nominal hourly wage. However, a worker's real wages are the wages adjusted for inflation. For example, if a worker's nominal wage rises by 2% and there is 3% inflation, we actually consider her real wages to have dropped by 1%--that is, the value of goods and services she could afford went down by 1%. (In fact, real wages did decrease in 2005: median weekly earnings increased from $631 to $651, an increase of 3.16%, but since inflation was at 3.4%, the $651 in 2005 was actually worth less than the $631 in 2004.)
However, the article doesn't say that real wages have decreased--indeed, with the exception of 2005, real wages have generally increased over the last few years. The article's main point is that as a percentage of GDP, real wages have been steadily declining. In other words, the growth rate of GDP has been steadily greater than the growth rate of wages and salaries.
But that's only part of the story, because wages are just part of workers' overall compensation packages. Health insurance is the other big component. Due to a historical anomaly, employers tend to pay for health insurance for their workers, rather than workers buying it on their own. Therefore, total compensation to workers can be thought of has having two components: wages, which the workers are free to spend as they wish, and a certain amount of money with which they are forced to buy health insurance. In recent years, the price of health insurance has grown at double-digit rates, far outpacing inflation, while wages, as a fraction of the overall compensation package, have been falling.
Note that the increase in health care spending might very well have had the same effect if workers received all of their compensation in cash, and then paid for health insurance out of their own pockets. Since health care is widely viewed as a necessity rather than a luxury, demand for it is relatively inelastic; therefore, as its price goes up, people might very well decrease their expenditures on other goods and services. In other words, the effect of an increase in the price of healthcare means a decrease in consumers' real income, and the fact that wages and salaries don't include benefits serves to highlight this effect.
However, even if we take into account both wages and benefits, overall compensation has still been falling in recent years as a fraction of GDP, while corporate profits have been increasing. Why? Workers have been more productive--that is, creating more output per hour of labor--which has meant higher revenues for firms. Those revenues accrue to different factors of production: wages to workers, rent to capital, and profits to shareholders. The article's main point is that the gains in worker productivity have not accrued to workers, but instead have gone largely to shareholders.
Describing how wages, benefits, and profits have changed over past years is the province of positive analysis. The article raises an important normative question as well: is it fair that wages have not kept pace with productivity? Or that the top 1% of earners received over 11% of all wage income, up from 6% three decades ago? These are important questions as well, but economic analysis is less helpful in answering them.
1. Suppose the U.S. government provided national health insurance for everyone. Do you think wages and salaries would still be falling? What about after-tax wages?
2. The divvying up of the economic pie appears to be a zero-sum game: there is a certain amount of economic surplus out there to be divided between profits and wages, and more of it has been going to profits. Jared Bernstein, a senior economist at the Economic Policy Institute, lays the blame for lower real wages on workers' lack of bargaining power. Would a greater rate of unionization help to achieve greater equity? Would it do so at the expense of efficiency? What about using government taxes and transfers to achieve greater redistribution of income from rich to poor households? Would economic policies aimed at a more equitable division of the economic pie mean a decrease in the size of the pie?
3. If labor markets were perfectly competitive, wages would increase directly with productivity. (For a simulation of how this works, see this demonstration of an interactive experiment based on the classical theory of the labor market.) What about the real labor market is a departure from this model?Update: This has been a hot topic on the blogosphere. Greg Mankiw, Russell Roberts, and David Altig have posted comments on the article (and others' comments on it). More to follow, I'm sure, and we'll keep you updated.