Real earnings crashed in the 1970s, continued to decline through the mid-1990s, and rebounded through the mid-2000s. The rebound is only about half of the earlier declines. Pearlstein wonders about why the last several years have been flat, given higher productivity growth over the last decade or so.
Here's annual productivity growth over a similar time period (with period averages prior to 1973, 1973-1996, and 1996-2007):
The periods of decline in the real earnings chart do correspond pretty well to the low period of productivity growth. But I don't think that's the whole story. Consider the following chart of labor force participation by sex over the same time frame:
The rebound in productivity also corresponds roughly to the end of the big increase in labor force participation. I'd suggest that the long-term increases in labor force participation, driven by women entering the labor force in larger numbers, also played a role in holding down real wage increases. With that trend having slowed or stopped, there was more scope for real wage gains as labor was compensated.
But adding this consideration of labor supply doesn't do anything to explain recent events--labor force participation has been lower during the same period that real earnings have been flat.
Some explanation can be found in the share of total compensation going to wages and salaries versus benefits. In March 2008, wages and salaries were 70.6% of total compensation in private industry. In March 2000 and March 1986 (the earliest report I could find), wages and salaries were 73% of total compensation. So the growth of benefits as a share of total compensation knocked 3 - 3.5 percentage points off the growth of wages and salaries over the last 8 years. Annualized, that translates into 0.5 percentage points per year, which is still small compared to the annual productivity growth rates.