August 15, 2003

Output and Hours Since 1960

Some more two-year centered moving averages since 1960: this time it's nonfarm business real output and nonfarm business total hours worked. Once again (and unsurprisingly) we are outside the bounds of previous experience in the modern American economy:

The little twig on the left side of the figure shows the moving-average data for 2000-2002. Never in any two-year period in the modern American economy's experience have hours fallen so fast. Given what has happened to hours in the recent past, the standard historical pattern would lead you to expect output to be falling at 2.5% per year or more--and you would expect productivity growth to be negative, not positive and in excess of 4% per year.

We are indeed in uncharted waters. Not that we should mind--extraordinarily rapid productivity growth is a wonderful gift. But it does pose different problems for economic management to solve than the ones we had gotten used to...

Posted by DeLong at August 15, 2003 10:43 AM | TrackBack


Well, rather than take advantage of the wonderful growth in productivity we have used absurd tax cuts to turn surplus to gaping deficit and allowed the economy to grow too slowly to produce jobs. I can not imagine a more stupid fiscal policy than ours these past 3 years. Of course, what does this Administration really care about jobs when the pretense of caring will suffice?

Posted by: Fairlt Balanced Anne on August 15, 2003 11:01 AM

I want to offer an anecodotally based suggestion for the productivity increase. I believe it's the availability of modern prescription mood altering drugs.

Both my grown children work for large companies, and they are being worked at rates that in the old days would have brought on nervous breakdowns, or at least degraded function. But they both take "calming pills" as do all there coworkers, so the press goes on.

Posted by: Dick Thompson on August 15, 2003 11:14 AM

Wish we could pin down how much of the productivity increases have been due to the pressure on labor rather than technical advance. Technical advance and application has been substantial, but there may well be a growing office push that is not easily accounted for.

Posted by: Fairesq Balanced Anne on August 15, 2003 11:32 AM

where does the hours worked data come from? is it likely to be measured in a stable manner over the past 5 years-- ie could we be seeing side effects of hours worked shifting into less well measured parts of the economy (such as from hourly workers to saleried workers going or from on-site to off-site work?)

Posted by: John on August 15, 2003 11:34 AM

How we explain ourselves -

To the Editor:

Within Bob Herbert's Aug. 14 column are examples of why economists are misunderstood by most people, and why savvy politicians can connect with their constituencies even while offering nothing more than empty rhetoric.

Criticizing the Bush administration's policies, the economist Robert Solow characterizes the tax cuts as "redistributive in intent and redistributive in effect." While his contention may be on target, there is nothing in his quote that the average person can easily understand.

President Bush's assessment of the American economy is decidedly more upbeat and comprehensible: "This administration is optimistic about job creation." Mr. Bush may be ignoring the reality of today's job market, but you have no trouble understanding what he's saying.

Smart politicians understand that if you want to wag the dog, first get people to nod their heads.

Powell, Ohio, Aug. 14, 2003

Posted by: jd F&B on August 15, 2003 11:36 AM

Going back to Larry Lindsey's railroad model and the railroad bubles of the 18th century, there are some parallels to today's economy. Laying railroad track required lots of workers and lots of investment and during the boom, tracks were overbuilt Similarly, we have built an internet infrastructure and laid more than enough fiber optic to run the system.

In the railroad days, the trains hauled the materials needed to build the railroads. Once the tracks were built, maintenance fell to much smaller section crews. During the internet building, people were needed to build the infrastructure, (web authoring tools is a good model) but once the structure was built, the workers required to maintain it are far fewer.

Early on in the railroad, there is not much profit because charges for freight hauling require the completed track. During early days of the internet, the online finance, billing, verification, etc had to be designed before the profits could be collected.

Early on with the trains it took a while to build up ridership and freight hauling. With the internet, it becomes more viable only as more people go online. If only 10 % are on line, then the technology can reach far fewer than if 70% are online. With something like the internet where most of the capacity is already built, more customers can be added without adding a lot more employees. Plus people have to learn how to use it.

Are we seeing a serious dislocation as an industry matures? The railroad bubbles burst and caused recession and unemployment. The economy did not recover until new parts of the economy emerged to take up the slack. The tech bubble has burst and coming out of the slump will require sectors of the economy other than tech to lead the way.

We are in a period where a new sector needs to emerge that can utilize productive investment. Perhaps what the economy needs is beyond the standard economic manipulation of money supply and fiscal stimulus and institute policy that speeds the development of the next wave that will lead the economy?

Another analogy is the mechanization of agriculture that led to the migration of thousands of farm workers off small plot farms and flooding the job market (As in Grapes of Wrath). Tech has mechanized the service world. ATM machines cash paychecks, not human tellers. Customers book travel direct, not travel agents, etc.

At some point the productivity growth has to level.

Posted by: bakho on August 15, 2003 11:50 AM

What if the employment answer were fairly obvious? What if we had increased government spending on America's infrastructure, rather than cutting taxes for the wealthiest? My Mom tells me there was no electricity in New York City last night. Could that be? No way....

Suppose we were spending more to build a more ample more reliable power grid, better public schools, more parks, better police forces for areas such as south Los Angeles. Suppose, suppose.

What if we were spending the sort of money that went for tax cuts to the wealthiest on public works for all?

Posted by: anne always Fair and Balanced on August 15, 2003 12:05 PM

We have been underinvesting in infrastructure for over 2 decades now. Infrastructure is important to Mfg location. This is why states and municipalities have to offer infrastructure incentives to get industry to locate large mfg facilities.

Maintaining infrastructure can be used to tread water, but we need the next innovations in energy, trasnportation and medicine.

Posted by: bakho on August 15, 2003 01:00 PM

Yes, yes. Building infrastructure allows more rapid more general growth. But, building to innovations in energy and transportation and medicine can be thought of as building infrastructure.

Will Calgary Be the Next Kuwait?

America has many reasons for its risky involvement in Middle East politics, but there is no question that dependence on imported oil plays a very large role. The fact is, however, that the largest deposits of oil in the world are not in the Middle East or in Russia or off West Africa or in the Caspian Sea area, for that matter. They are in two Western Hemisphere countries: Venezuela and Canada.

The catch is that they consist of bitumen and other forms of "heavy oil" petroleum buried in sand deposits that requires a special refining process to be converted into so-called syncrude. This conversion is expensive, but the end product has all the favorable qualities of conventional light oil. And these heavy oil deposits occur in mindbogglingly large quantities....

Posted by: anne F&B on August 15, 2003 01:08 PM

Could someone please tell me how ow are ours worked measured?
If it's merely official ours worked, then doesn't this graph simply show that fewer people than ever are being paid by the our?

It seems to me that all that's happened is that business risk has been shifted from investors to employees. Hence in a downturn, bosses can fire half their workers and have the other half work longer hours to make up the difference.

Hardly seems like something to celebrate.

Posted by: WillieStyle on August 15, 2003 01:13 PM

I don't really get bakho's argument here as both anne and WillieStyle's posts and the discussion in general seems to be all about faltering demand. How could the 'next invention' stimulate demand - doesn't inventions add to productivity hence further cooling off labour market and cons.conf?

And according to Virgina Postrel many 'next inventions' has already been around for a while, like the nail salon:

Posted by: Mats on August 15, 2003 01:55 PM


Compare the demand for iPOD today with demand 10 years ago. (No product, no demand)

Posted by: bakho on August 15, 2003 02:19 PM

How about training workers in India to do your job so you can be laid off? It increases productivity if the output of the foreign workers is counted as an output cost.

Posted by: bakho on August 15, 2003 02:21 PM

OK, bakho, "(No product, no demand)" and I don't think not even I could miss your point now. But Postrel's argument is that we already have lots of new products and services, like the nail salon and "aestethics". We have the iPod and the SUV and what not.

Yet demand falters: No job security, no demand.

Posted by: Mats on August 15, 2003 02:36 PM

That graph is kind of hard to get anything out of. Do you have one with a relationship productivity/hours and time?

Posted by: northernLights on August 15, 2003 03:37 PM

Bakho's comments touch upon some questions that I've been mulling over.

Where does R&D fit into the picture?

Suppose a company has 10% its staff dedicated to R&D. Laying them off would have no (short-term) affect on output. Would this alias as a productivity gain? Similarly, when a start-up company with no revenues goes into 'hibernation' to preserve cash does this look like an increase in productivity? What about not forming a startup company at all? In each case there's no change in short term output but there is a reduction in hours worked.

Running along the same lines, what happens when there are layoffs in an industry with overcapacity? Does this boost productivity as well?

If the answers to these questions are yes, then unwinding an investment bubble would boost productivity. How large would an investment bubble need to be in order for its unwinding to have a meaningful effect on the overall productivity of the economy?

Although it's relatively small compared to the overall economy, venture capitial went from ~100 billion in 2000 to ~15 billion now. I'm too lazy to put numbers to the telecom bubble, much less other sectors, today. But it doesn't take to many 100 billions to add up to real money.

Posted by: chris_a on August 15, 2003 04:40 PM

Brad DeLong points out that the deficit of the 1980's appears to have constrained productivity growth. Low saving, low investment, low productivity growth. Well then, we will soon be at ZERO national private and public saving. Will the deficit tell on productivity?

Posted by: anne on August 16, 2003 09:31 AM
Post a comment