Productivity was revised down for the first quarter to an annual rate of 2.8% from an original report of 3.6% growth. In this post, the current numbers are presented first, and the original numbers are shown in parenthesis, so you can see exactly where the overall revisions came from.
Output rose 4.0% (4.4% before the revision) while hours worked rose 1.1% (0.8%) on a seasonally adjusted annual rate basis. The downward revision was greater than expected; the consensus was for productivity to be revised down to 3.3%.
It is not all that surprising, given the downward revision to GDP growth numbers in the first quarter. After all, economic growth can be broken down into the change in the total number of hours worked multiplied by the change in the amount of output per hour.
Productivity is just another term for output per hour. Thus, if there was less growth in output (GDP) than originally thought, either there had to be fewer hours worked (i.e. fewer people working) or there had to be less of an increase in output per hour.
The 2.8% growth level in productivity is still pretty solid, especially as it comes on top of huge gains in the second half of 2009. Year over year, productivity increased an incredible 6.1% (6.3% pre-revision) as output increased 3.0% (3.1%) while hours fell 3.0% (also 3.0% before the revision).
The year-over-year gain in productivity was the largest since a 7.0% increase in the first quarter of 1962! Since 1948 (when the data starts), the 6.1% level of productivity growth has only been exceeded six times. The first quarter was a significant slowdown from the fourth quarter productivity pace of 6.3%.
Manufacturing Productivity
Generally, manufacturing productivity tends to rise more (but also be more volatile) than overall productivity. Large sectors of the service sector have historically been almost immune to productivity increases, and government — literally by definition — cannot become more productive. Even if magically one FBI agent were able to do the work of the entire Bureau and do it just as well, it would still not show up as in increase in overall productivity in the GDP numbers, as the government’s contribution to GDP is counted by input rather than output in the GDP accounting.
These numbers just refer to non-farm businesses. Still, there are very real limits on how much productivity can increase in many service industries. For example, there has been no real increase in the productivity of barbers in about a century, as the amount of time it takes to cut someone’s hair really has not changed. Health care is one larger sector that has historically been immune to higher productivity as well, or at least the increases there have not been as well measured.
The first quarter was somewhat of an anomaly in that regard, as manufacturing productivity rose by just 1.5% (2.5% before the revision) but it is up 7.2% (7.5%) year over year. Manufacturing output in the quarter was up a sharp 7.2% (7.5%), but hours worked rose by 5.6% (4.9%, in other words, employment was revised upwards). On a year-over-year basis, output is up 3.2% (3.3%) and hours worked are down 3.0% (3.9%).
Given how much smaller manufacturing is as a share of the overall economy now than it was in the 1950’s or 1960’s when we last saw these sorts of overall productivity gains, the current increase is especially impressive. Durable goods manufacturers like Ford (F) and Whirlpool (WHR) were able to continue increasing productivity at a faster rate than the economy as a whole, with growth of 3.0 (3.8%) on the quarter as output surged 10.2% (10.5%), while hours worked increased 7.0% (6.4%). On a year-over-year basis, durable goods manufacturing productivity is up 8.7% (9.0%) as output is up 3.0% (3.1%) while hours worked are down 5.2% (5.4%).
It has been non-durable manufacturing where productivity gains lagged in the quarter, with productivity up 0.9% (2.2%) on the quarter with a 3.3% (4.8%) gain in output and a 3.4% (2.5%) increase in hours worked. On a year-over-year basis, non-durable manufacturing productivity is up 4.7% (5.0%) on a 3.3% (3.4%) increase in output and a 1.4% (1.6%) decline in hours worked. It is not surprising, then, that the earnings gains for non-durable manufacturing firms such as Kellogg (K) and Altria (MO) have not seen the same sort of pop that more cyclical durable goods firms have (nor did they have the implosion in earnings earlier).
Post-Recession Resurgence
Productivity tends to rise coming out of recessions. Employers are able to increase production without having to increase the number of workers right away. Actual declines in productivity are relatively rare, with only 28 quarters (11.2%) recording year-over-year declines since 1948, and ten of those were between the second quarter of 1979 and the third quarter of 1982. However, going into and in the early stages of recessions the rate of productivity growth does tend to slow dramatically.
Over the long term, there is probably no more important economic number than productivity. It provides the key to per capita GDP, or how rich a country is. If output per hour is stagnant, the only way that the economy can grow is by adding more hours, which means adding more people.
Right now, with unemployment at 9.9% (well at least for April it was; we will see tomorrow where it was for May, but it is expected to remain unchanged or dip only slightly even as we add 500,000 more jobs, mostly due to the Census), adding more hours might be better. However, in the long run you really only add hours by adding more people.
GDP per Capita & Unit Labor Costs
It really is GDP per capita that matters, not total GDP. After all, the GDP of India ($3.53 Trillion, IMF on a PPP basis) is substantially higher than the GDP of Germany ($2.81 Trillion), but nobody would think that the average person in Mumbai is better off economically than the average person in Munich. The difference is that while there are many more Indians than Germans, each German produces more. If productivity goes up, then it is possible to pay the workers more.
But just because something is possible, it does not mean that it has to happen. Historically, gains in productivity have been split between capital — in the form of higher profits — and labor — in the form of higher wages. Capital shares in the productivity gains because it is better capital that provides the bulk of the productivity. It is not that the Germans work harder each hour they are working than Indians do, or that they are inherently smarter. The difference is that they have better tools to use, and better infrastructure that cuts down on waste.
More recently, though, productivity has been surging but wages have not been going up very much, and the vast bulk of the gains have been going to capital. Unit labor costs fell 1.3% (1.6%) in the first quarter — the 2.8% (3.6%) increase in productivity outpaced a 1.5% (1.9%) gain in wages. Unit labor costs fell 4.2% (3.7%) year over year, as the 6.1% (6.3%) increase in productivity outpaced a 1.6 (2.3%) rise in pay.
The decline in how much it costs an employer to make something, or provide a service, is a big part of the reason that earnings have been growing so fast. On a year-over-year basis, total net income grew 49.7% in the first quarter for S&P 500 firms.
The graph below shows the year-over-year change in both productivity and unit labor costs over the last 40 years. While there is a tendency to have the lines move in opposite directions, that tendency has become much more pronounced in recent years.
Falling unit labor costs are actually fairly rare. While the year-over-year decline in the first quarter was less than the 4.6% year-over-year decline in the fourth quarter, with that exception it would have been an all-time record decline, with the next closest decline being a year 3.5% year-over-year drop-back in the first quarter of 1950.
Since 1948, only 33 quarters, or 13.3% have seen year-over-year declines in unit labor costs, and only 19, or 7.6% of the declines have been greater than 1.0%, and both of those numbers include the last three quarters, which have all been deeply negative. Falling unit labor costs are far more common for manufacturing than for the economy as a whole, with declines in 36 (40.4%) of the 89 quarters since the data has been kept.
The Pros & Cons
For stocks, the increase in productivity and the fall in unit labor costs are very good news as they are the key to continued profit growth. For the economy as a whole, it is a bit of mixed bag. Being able to produce the same output with fewer workers is not the way to solve the unemployment problem. The rise in productivity will increase the average level of income in country, but with plunging unit labor costs it is also exacerbating the already extremely high level of income inequality in the country.
Averages, after all, can be very misleading. If Warren Buffett walks into a bar, on average, all the people there are billionaires. Over the long-term, ever-increasing shares of GDP going to capital rather than labor is not going to be sustainable. There is a limit to the numbers of cars and refrigerators the very rich are going to need. If the average income goes up, but it is all concentrated in a few hands, then there will not be that much demand for many products.
Without that demand, output and sales will start to fall, which will eventually impact profits. However, in the short term, the rising productivity and the falling unit labor costs are very good news for the bottom line. If the benefits from rising productivity were more widely spread, it would also be very good news for the economy as a whole.
Dirk van Dijk, CFA is the Chief Equity Strategist for Zacks.com. With more than 25 years investment experience he has become a popular commentator appearing in the Wall Street Journal and on CNBC. Dirk is also the Editor in charge of the market beating Zacks Strategic Investor service.
