In the second quarter, the economy was hemorrhaging jobs, losing a total of 1,265 million. While that was not as bad as in the first quarter when it lost 2.074 million jobs, it was still pretty awful. However, in the first quarter the economy as a whole shrank by 6.3% (annual rate) while in the second quarter the contraction slowed to just a 1.0% rate.

Well, GDP measures output, and output is a function of two things: the number of hours worked and the output per hour. Output per hour is also known as productivity, and today’s productivity report shows the reason for the greatly improved (less bad) performance of the economy.

Non-farm business productivity surged 6.4% in the second quarter, well above the 5.5% rate the Street was expecting. This was the biggest increase in productivity since the third quarter of 2003. That is a fairly interesting date, since it was the point at which the economy really started to recover from the previous recession (even though it officially ended in late 2001, it still felt like a recession through most of 2003). The surge in productivity also helps explain why so many companies have come in with much better-than-expected earnings.

Manufacturing productivity grew 5.3% in the quarter. This happened because hours worked fell by 14.4% (not only fewer workers, but the remaining ones working fewer hours) while output fell by “only” 9.9%.

In durable goods, manufacturing output fell 16.5% but hours plunged by 19.6%, for a 3.9% productivity gain. Output of non-durable goods was more stable, falling 3.4% in the quarter, but hours worked fell by 5.3%.

For the non-farm business sector as a whole, output was down 1.7% while hours were down 7.6%.

Generally speaking, you want to see productivity rising. It is the only path towards raising income per capita on a sustained basis. However, you would prefer to see it happening from output rising faster than hours worked rather than from hours worked falling faster than output.

Over the long term, economic growth is the combination of population growth and productivity growth. When a country is operating below its potential (a.k.a. being in a recession), the working population can grow even if the total population is unchanged by increasing the percentage of the population that is working (in many ways the employment-to-population ratio, now down at 59.4%, is a more important statistic to look at than the unemployment rate).  Still over the long term, the level of productivity is what determines living standards in a country.

Sweden’s economy is smaller than that of Indonesia, but nobody would say that Indonesians are richer than Swedes. The output per hour of Sweden dwarfs that of Indonesia. The only reason that Indonesia has a bigger economy is that its population is an order of magnitude bigger than that of Sweden. Thus overall, productivity is a vital measure of the economy, and growth in it is a good thing.

However, when productivity rises due to falling hours rather than rising output, it can lead to greater disparities of income in the country. The people who are out of work get less income, while businesses remain profitable.

Indeed, the other part of the report today was on unit labor costs, which are plunging. For the non-farm business sector as a whole, real compensation per hour (not just wages but also including benefits) fell by 1.1%. By keeping output up and cutting hours, this meant that unit labor costs fell by a stunning 5.8%.

Since for many (most?) businesses labor is the most important cost, falling unit labor costs can greatly boost a businesses’ productivity. The improvement in unit labor costs was not as great in manufacturing as it was for non-farm business as a whole, so presumably we are seeing a huge improvement in service sector unit labor costs.

In manufacturing, unit labor costs were up 0.5% due to a 4.4% rise in real hourly compensation. That is a surprisingly high number, and in durable goods real compensation was up 7.3%, causing unit labor costs to rise 4.7%. I will have to investigate what special factors, if any, led to such a jump in real compensation in the face of spiking unemployment. Non durable real compensation and unit labor costs were far more subdued at 0.7% and 0.0%, respectively.

Part of the reason for the unexpectedly strong reading in second quarter productivity is that it was coming off a lower base than we thought, as the productivity numbers for the first quarter were revised down to a gain of just 0.2% from a previously reported 1.8%. That is because output fell by 8.7% in the first quarter, not by the previously reported 7.2%. Hours worked were revised only slightly to a decline of 8.9% rather than 8.8%.

However, real hourly compensation was dramatically revised to a decline of 0.1% rather than the previously reported increase of 7.0%. The lower starting level there helps in part explain the big jump in real compensation in the second quarter, particularly in manufacturing, where the previously released number for the first quarter was an increase of 16.1% in real hourly compensation and a rise in unit labor costs of 16.6%. Those were revised down to just increases of 4.6% and 4.9%, respectively.

Given the size of that revision, it does not give one that much confidence in the second quarter numbers that were just released. The productivity numbers for all of 2008 were also revised downward to growth of just 1.9% from the previously reported 2.7% increase. That revision was entirely due to a downward adjustment in the estimate of output, to unchanged from 0.8% with hours unchanged at a decline of 1.9%.

While this sort of growth in productivity is unlikely to last, and it is largely a function of people getting laid off and the remaining work being spread among fewer people, it is still overall a good thing for the economy.

It is most certainly a good thing for companies. The surge in productivity goes a long way towards explaining how companies as diverse as Colgate-Palmolive (CL), McDonald’s (MCD), International Business Machines (IBM) and Fluor (FLR) could manage to report higher earnings than a year ago in the face of falling revenues.

Negative operating leverage is normally a very rare thing, but was surprisingly common this quarter, with a total of 77, or 17.0% of the 452 S&P 500 firms that have reported have enjoyed higher EPS than a year ago, even though their revenues were down.
Read the full analyst report on “CL”
Read the full analyst report on “MCD”
Read the full analyst report on “FLR”
Read the full analyst report on “IBM”
Zacks Investment Research