The chart above is not a "serious" statistical analysis, but it tells us that there is a linkage between the growth rates of real GDP and output-per-hour. A visual inspection shows that the trends in the 5-year growth rates are similar, although the correlation is not perfect. In particular, one could note that the 5-year annualised growth rate of real GDP has picked up, while productivity has not, and so one could attempt to argue that the current productivity slowdown is not just the result of weak growth.
Nevertheless, I would stick with the slow growth explanation. It matches my intuition about the business sector. In my view, most firms cannot adjust their labour forces to match the demand that they face. A classic example would be a restaurant; if very few diners walk in that night, the restaurant will have a lot of under-employed waiters and cooks. If the situation persists, the restaurant would eventually drop the number of employees to match lower demand. But in most cases, such adjustments will be slow, as it is costly to hire and fire workers. More generally, a good portion of work force represent fixed overhead, and so productivity drops if demand is less than expected.
(Update: Commenter Ramanan notes that this is more formally known as the Kaldor-Verdoorn Law in economics. Link to Wikipedia article for Kaldor's growth laws, and Verdoorn's law.)
There are some exceptions; firms with long-dated contracts or large orders (plans) can adjust their hiring to match those contracts. Also, firms like oil drillers can have a well-defined output per employee in the field. But at the same time, those firms have administrators that cannot be adjusted to match the wiggles in demand.
Therefore, I would expect that productivity would follow growth over shorter periods of time, such as a business cycle. The recent shortfall in productivity versus growth is not that dramatic (growth is still weak), and could easily be explained by a shift in the sectoral composition of the economy.
In the short term, I doubt that productivity tells us much about inflation. Even if productivity is weak, higher wages could coincide with weak inflation in consumer goods prices; the profit share of income would bear the brunt of the adjustment. (This recent article touched on similar topics.)
Longer-Term ProductivityEconomists tend to wring their hands about long-term productivity. Structural trends should trump the cyclical factors I discussed above. (To take my restaurant example, the average capacity utilisation in the restaurant sector should tend towards some average level over time, possibly by the elimination of weaker restaurants.)
I doubt that economics as a field has anything useful to tell us about productivity. In the short term, we could help matters by running the economy closer to full capacity, which was done in the early post-war era. But once full employment is achieved, this would provide no further gains.
We then need to turn to technology. The huge advances in productivity in the twentieth century were mainly associated with the adoption of technologies revolving around hydrocarbons. (This includes artificial fertilizers derived from natural gas.) Additionally, as someone trained in electrical engineering, we need to include improvements in harnessing electricity and electronics. Industrial farming techniques allowed for the huge migration from rural areas to urban.
However, the low-hanging fruit have been picked. Modern electrical engineering largely consists of finding the optimal way of distributing cat pictures. Although there is an insatiable demand for cat pictures, people do not want to pay very much for them. This means that paid output is stagnant, even though "consumer utility" might be rising.
With oil prices collapsing, it may seem strange to worry about resource limitations. Nevertheless, I still believe that resource limitations will force society to move backwards on its physical productivity; whether this is captured by GDP measures is uncertain.
(c) Brian Romanchuk 2015