Funny, I was just in a discussion with some folks on this very issue, so enjoy the citation bomb. In short: total compensation isn't outstripping productivity by a substantial degree.
First, start with "The relative stickiness of wages and prices" by David E. Spencer. http://findarticles.com/p/articles/mi_hb5814/is_n1_v36/ai_n2... It provides important background to the discussion of whether wages or prices 'stick' more, and provides an overview of the debate through '98.
Next, we turn to Martin Feldstein's paper, "Did wages reflect growth in productivity", 2008. http://www.nber.org/papers/w13953 He highlights two common errors with studies that find wages rose substantially slower than productivity. First, he explains that such studies focus on wages rather than total compensation. Without comparing productivity to total compensation, such studies are dangerously flawed. Second, such studies tend to poorly select the deflator when comparing productivity & compensation. Basically, the price index that's used to determine compensation is different than the price index used to calculate productivity.
For an even more recent discussion of the same problem, but within the context of Russia's economy, read "Productivity and Cost of Labor: How Statistical Illusions Are Born" by R. Kapeliushnikov, May '10 http://mesharpe.metapress.com/index/BW00W218X2143504.pdf. He explains:
It seems that many of them are unaware of the difference between the
producer real wage and the consumer real wage. The former is estimated
by deflating the nominal wage using the producer price index (PPI) or
the GDP deflator; the latter, by deflating the nominal wage using the
consumer price index (CPI). One reflects the change in the price of
labor from the perspective of firms; the other, the change in the pur-
chasing power of wages from the perspective of employees. Therefore, if
we want to answer the question of how expensive or cheap labor is for
firms, we should use the producer real wage, not the consumer real wage.
Hopefully this helps in evaluating the claims made by the EPI (the source of the data used by the NYT). I'm not familiar with any direct responses to their latest data, but Feldstein reaches a different conclusion over much of the same time period: While there's some lag in total compensation, it's not dramatically different than the rise in productivity.
We're talking about the first graph, right? If your "statistical artifact" point is to have any merit, then you would have to demonstrate that something changed significantly in how those statistics were obtained after 1980 - because that's when the behavior of those time series changes significantly. If you cannot do that, then the other theory - namely, policy changes - seems the far more plausible explanation.
First, start with "The relative stickiness of wages and prices" by David E. Spencer. http://findarticles.com/p/articles/mi_hb5814/is_n1_v36/ai_n2... It provides important background to the discussion of whether wages or prices 'stick' more, and provides an overview of the debate through '98.
Next, we turn to Martin Feldstein's paper, "Did wages reflect growth in productivity", 2008. http://www.nber.org/papers/w13953 He highlights two common errors with studies that find wages rose substantially slower than productivity. First, he explains that such studies focus on wages rather than total compensation. Without comparing productivity to total compensation, such studies are dangerously flawed. Second, such studies tend to poorly select the deflator when comparing productivity & compensation. Basically, the price index that's used to determine compensation is different than the price index used to calculate productivity.
For an even more recent discussion of the same problem, but within the context of Russia's economy, read "Productivity and Cost of Labor: How Statistical Illusions Are Born" by R. Kapeliushnikov, May '10 http://mesharpe.metapress.com/index/BW00W218X2143504.pdf. He explains:
Hopefully this helps in evaluating the claims made by the EPI (the source of the data used by the NYT). I'm not familiar with any direct responses to their latest data, but Feldstein reaches a different conclusion over much of the same time period: While there's some lag in total compensation, it's not dramatically different than the rise in productivity.