December 04, 2002
Six Questions About Productivity Growth
Six Questions About Productivity Growth
Posted by DeLong at December 04, 2002 03:11 PM
- Rapid American productivity growth has continued through the recession. What conclusions should we draw from this?
- The differences in productivity growth between the U.S. and mainland Europe--especially in infotech-heavy services. Why, and what does it mean?
- The current intellectual property wars. How will they be resolved? What implications do the intellectual property wars have for economy-wide productivity growth.
- The size and salience of our current industrial revolution. How does it compare to past industrial revolutions? What are its social and political implications?
- The developing world. What does the forthcoming transformation of the service sector into a tradeables sector mean for the world economy?
- How fast is economic growth? How large are the biases in our standard statistical measurements?
I long ago concluded that Robert Solow was incorrent
regarding the impact of technology on our national economy. My guess is that we being rewarded by the huge numbers of our citizens who are very computer savy. Solow didn't understand that the rewards are often subtle and not immediately obvious. The Europeans are behind the curve. What's new about that? I'd say more but some people will accuse me of unfair Euro bashing.
Oh yee of little faith in our public institutions.
We will work out the intellectual property issues.
This will be old hat in a few years down the road.
Industrial revolution? Have we already forgotten that just a hundred years ago around half of our economy was devoted to farming? The same will eventually hold true for manufacturing.
The Third World requires stable political institutions. No one is starving today because of economic reasons! Little can be done to assist these people until the Castros, Mugabes, Husseins, and other monsters are removed from power.
It's worth making the distinction between economic output and labor input. 100 years ago half of the American population may have lived on farms. But with the mechanization of farming, the industrial revolution and improved planting techniques the nation was able to dramatically increase agricultural output even while the number of farmers decreased.
With manufacturing, the proportion of GDP accounted for by manufacturing hasn't decreased; but the proportion of the workforce devoted to manufacturing has fallen sharply.
Funny how much media attention is devoted to aggregate data such as GDP, when it's really GDP per capita that matters to national well-being. The intangible, unquantifiable driver of productivity growth is information and communication. I used to follow NCR closely back before ATT bought the company and ruined it. NCR had the highest return on invested capital of any computer company in the business - and management explained it by noting that they worked hard at not producing any more of a product than they had demand for - so they never had bloated inventory, too much returned product, etc. Writ large throughout the economy, the internet and sophisticated software ought to be able to help everyone accomplish much of what NCR did through true grit back in the good old days - before Al Gore invented the Internet.
1) What role are hedonic adjustments playing in the difference between US and European output and hence productivity. Correct if wrong, but US data are much more prone to hedonic adjustments and the European Data (certainly German data) are not. Guessing that plays a big role ...
2) Hours worked ... any chance the hours worked (lap tops / home / cell phones) etc. just arent captured in the traditional commerce and bls statistics ?
If one were looking for the primary difference between Europe and the US in computer usage by individuals, then one must take into account the ability of Americans starting in the early 90's to gain internet access via dial-up at a flat rate. The flat rate that is important in this case is the flate rate for local phone calls in the US. European phone systems were almost exclusively billed per minute for usage, making even flat rate internet services prohibitively expensive for larger numbers of hour usage.
As a result, American's have had the benefit since the mid to late 90's of a large pool of younger people being exposed to cheap internet access. If a youngster was so inclined they could become a productive member of society at 16 years of age, due to personal interest and exposure to hundreds of hours of network time. Even more casual users have been exposed to common tools and formats of usage of network services.
This inexpensive exposure undoubtedly has helped to contribute to the vast pools of labor with at least basic understanding of computers and networks, not to mention providing an additional reason for individuals to purchase computers for home use and become involved with computers and networks.
Pretty big advantage to be gained from most areas having flat rate for local phone calls.
Did I mention that a lot of local phone services are trying to move to per minute charges for local calls? An area near us is paying 2 cents per minute for local calls. 60 hours a month of internet access would result in additional $72/month in phone bills.
Gee. That Robert Solow. Wrong again, huh?
Nah, he was probably right, at least in that the immediate effects of computer technology didn't translate into an immediate productivity boom. The long-term effects are another matter.
But think about it. What's new IT and communications technology going to do to our production possibilities?
- INFORMATION: A big part of the internet is about access to information. I can find out stuff in minutes I never could have before. But how much of that actually increases total productivity? I mean, I like being able to go to IMDB and see who starred in what film, but that's a consumption benefit, not a productivity enhancement.
- TRANSACTIONS COSTS: So it's now easier for me to order books or DVDs that aren't in my local retail outlets. Again, that's great, and it opens up the possibilities for online retailers to emerge. But they're still competing against bricks-and-mortar establishments with localised advantages. There's some improvement in economic efficiency, but it might be 2nd-order or 3rd-order in magnitude.
- PRODUCTION TECHNOLOGY: Computers allow automation and hence improved production processes. But this process has been going on for decades and there's no reason to think the personal computing boom would have led to a HUGE increase in the trend here. And internet/intranet developments might lead to better coordination in multi-plant operations, but again, what's the empirical significance here?
I love the internet and get a whole lot of benefit from it. But office-time spent web-surfing and posting to blogs doesn't increase my measured productivity, in fact, more likely the reverse. There's a lot of CONSUMPTION benefit to be accounted for before we start assuming all sorts of PRODUCTIVITY improvements.
"I can find out stuff in minutes I never could have before. But how much of that actually increases total productivity? I mean, I like being able to go to IMDB and see who starred in what film, but that's a consumption benefit, not a productivity enhancement."
I strongly disagree. Please give this a bit more thought. The consumer benefiting today should become a better producer tomorrow. You are also able to research career related items. Gosh, think about how much you learn visiting this web site. Me thinks that you take your dramatically increased ability to access information for granted. Andy Grove, the former CEO of Intel, was even able to research the Internet to learn how to cure the cancer threatening his life. The world would have lost many years of his brilliant mind had he prematurely left this vale of tears.
I observed a young lady a short time ago who was not particularly brilliant. Nonetheless, she effortlessly could make good use of the Microsoft Office suite. There are many like her in the United States. This is not the case in most areas of the world. I am at this very moment employing a simple word processing program to assist me in composing this short response. Are you old enough to remember using a typewriter? Need I say more?
Robert Solow may have had a decent argument when many computers sat on desks merely collecting dust. A number of snobbish executives quickly lost interest and begged their secretaries to return to work. Our nation had to undergo a major learning curve. That was some ten to fifteen years ago. It is now 2002 AD. The younger executive are today writing their own e-mails. Therefore, the disgustingly cruel gods of creative destruction demanded that the secretaries be fired. Damn those bastards!
I think it's because the Canadians and Europeans parasite off us. No question about it.
You missed the bit where I said: "Nah, he was probably right, at least in that the immediate effects of computer technology didn't translate into an immediate productivity boom. The long-term effects are another matter."
In response to "I am at this very moment employing a simple word processing program to assist me in composing this short response" -- first, yes I remember typewriters. I love the immediacy and convenience of word-processors, the ability to edit, copy, paste, all that stuff.
But I remember the remark of a professor who was educated in the typewriter era, and supervised grad students in the word-processing era. Now, he said, they do draft after redraft, constantly tinkering and re-arranging and fussing and fixing. In my day, after prelimiinary hand-drafts, we had to write a really tight complete first draft, get it typed up, get it commented on and corrected, and get the corrected version re-typed. Then submit.
Dithering and faffing about (a friend's favourite term) has become much less costly, and so, as usually happen when things become less costly, more of it happens. There's LOTS and LOTS of extra paper output now because it's so easy to produce. But it's a stretch to say our national PRODUCTIVITY has immediately gone through the roof.
I love computers, I really do. In that context, I think it very important not to overhype the crap out of their productivity benefits. Let's keep it in perspective.
And I assure you, I have given this quite a bit of thought.
I should add that, as I understand it, Solow's comments derived from a period when the productivity impacts of computers were NOT showing up in the stats. In otherwords, despite, as he said, all these apparent productivity improvements like word-processing and so on, there was no empirical evidence of major AGGREGATE productivity gains.
That situation has changed somewhat since then.
So, at the time, he was hardly "wrong". He was commenting on the ecidence as it was at the time.
>>The size and salience of our current industrial revolution. How does it compare to past industrial revolutions? What are its social and political implications?
Perhaps the really interesting question here is WHERE is it taking place. We've had one in the UK, one in the US, it couldn't be China's turn now, now could it?
"So, at the time, he was hardly "wrong". He was commenting on the evidence as it was at the time."
Yeah, I think that my position concerning Solo is indefensible. I can't honestly criticize him while also stating that the computer was often no more than an expensive paperweight when first introduced into many offices. Oh well, I guess I will have to get on my knees and beg Solow for forgiveness.
Word processing has indeed made it easy to put out a lot of junk. One could always claim that making a task easier results in a lot mediocre work. We should probably bring back the horses and abandon our cars. Think about all the family dinners not eaten because everyone is going off in their own separate ways. What happened to our sense of community? Where’s Ned Lud when you need him?
"Perhaps the really interesting question here is WHERE is it taking place. We've had one in the UK, one in the US, it couldn't be China's turn now, now could it?"
This is a question that only China's oligarchical leadership can answer. Are they willing to give up their power? You cannot have a vibrant economy unless people are free to use today's technology.
Productivity is economics without prices. 'Nuff said?
Actually, I think his/her comments are quite interesting. But 3.5 years too late, I'm determined not to permit even harmless jokes about a particular political slander go by untouched. Al Gore did not claim to invent the internet, and what he claimed to do he, in fact, did. If nothing else, you should footnote the RNC when you quote them.
I can't believe it. I agree with David. His first post comments on the need for better institutions. The bedrock of economic growth is not technology, its institutions.
Without proper institutions growth doesn't happen. David later points out that China's oligarchical leadership must change to enable China to realize its potential. Once again, institutions.
Look at a country like Nigeria. Valuable nautral resources, large number of educated people, crappy institutions.
The problem is a country's institutions develop from its culture. How do you induce a developing country to truly embrace cultural change? That's the tough part.
JRoth: It was a joke. T'was a time when Dem's, Lib's, Repub's and Conservatives could laugh at themselves and at each other. I too know that Al Gore never claimed to have invented the Internet . . . . . at least half of the big scandals on every side of politics are partly true and mostly fabricated.
Anyway, it's not just the creation of institutions but the support and maintenance of existing ones that helps productivity growth keep marching along. Beyond the troubles of resource rich but growth poor nations such as Nigeria and Saudi Arabia are the awful disintegrations of formerly productive places such as Argentina.
Political thinkers and economists in Europe ought to be examining what's happened in Argentina over the past 40 years and trying to figure out how to keep Europe from traipsing down the same path in a decade or two. But they won't.
"I can't believe it. I agree with David."
To know me is to love me. Or at least, that's what my objectively unbiased mother says. There is no reason, for instance, whatsoever for anybody to go to bed hungry. Malnutrition is primarily a political problem. We long ago solved the practical challenges. This is why I get so upset hearing folks exaggerating the need for financial donations. It is far more important to change the backward cultures and corrupt political systems of these pathetic nations. Does anybody in their right mind really wish to provide added funds to a Robert Mugabe in the bizarre hope that his citizens will not starve to death? Liberal political correctness hinders our ability to confront the real reasons for poverty. One is not suppose to criticize the less than ideal political leadership of The Third World countries.
"Where’s Ned Lud when you need him?"
His spirit lives on at http://ludd.net... David, I'm sure you'll dig it.
Ha ha! Do you think David Thompson realises the irony of this statement
"Word processing has indeed made it easy to put out a lot of junk"
This doc says the 4.0% (now it's at 5. something) was revised to show only 3.1% Gross domestic product
If productivity growth is really as good as our government says it is why did Yellow Freight Lines go belly-up, Consolidated AND some other large trucking company (don't remember the name right-off-hand) filed bankruptcy. Doesn't productivity have to be shipped?
Isn't it possible that with all the lay-offs someone is "embroidering" about the amount of productvity the US actually had? Remember that IBM laid-off 15000 people and several other companies laid-off large sumes of workers too (WorldCom). How do you increase production with that many employees missing? Isn't that a physical impossibility?
I’m sorry, I’m not an economist but I don’t see how all that “productivity” helped us because I tell you folks, I just don’t see it in newspaper headlines as a plus with all the other bad news in the business section these days.
"Yeah, I think that my position concerning Solo is indefensible."
No, David, just confused.
Word processing aside, there are real productivity gains from IT. A few examples: real-time inventory reporting and control, automatic order placement, just-in-time delivery. All of these are enabled by IT technology. Today, a large business is able to realize efficiencies in economies of scale that were simply unimaginable just a few years ago. Managers can act on information that they were not even able to generate previously. It is really astounding.
To begin to take on some of Brad's homework: The social and political implications of IT advances are huge. It will be an entirely different society.
1) What Poindexter is contemplating is bleeding edge information processing. The amount of data that he wants to sift is astounding --that's why his budget is hundreds of millions of dollars just for the investigation (if I remember correctly). He has to buy super computers and tons of storage and that's just to test the proof-of-concept.
2) There is a new age of corporatism being ushered into being built on the same IT infrastructure. Corporations will know even more about you than Poindexter will (remember opt-out?) MTV's feedback loop is days or weeks already. They'll be selling your daughter a knock-ff of J-Lo's Grammy Awards dress before she even wears it to the show. Think of the marketing possibilities --if you think pop-unders are annoying, just wait. It will be targeted marketing everywhere you go. Perhaps AT&T will track you more closely through your cell phone and credit card with less restrictions than the FBI. Add things like TiVo tracking and Experian will know you better than your mother does.
The use of corporate data is limited only by the imaginations of the owners of the data (that's not you). The privacy implications are really enormous. I think most of the privacy advocates have already given up in the face of AOL-Time Warner, Disney, and Microsoft money. Our elected representatives certainly have been rendered mute, so it's all there for the taking.
Cheryl T: a famous economist once wrote about the "creative destruction" of capitalism. Strong economies and productivity growth do NOT guarantee that businesses will prosper; on the contrary, they guarantee that some businesses will FAIL. but the ones that fail are the inefficient, less productive businesses and the ones that prosper are the less inefficient, more competitive businesses. in an economy where lots of businesses DON'T fail, you end up with a bunch of companies run like the US Postal Service.
Sure, IBM has laid lots of people off over the past 10 years, but its sales have grown . . . . the corporate equivalent of productivity growth is "sales per employee". IBM's sales per employee have been growing smartly - not necessarily because the numerator has been expanding rapidly, though. And on average, the people IBM has laid off got jobs elsewhere, that's why the aggregate employment data in the US has been growing steadily.
a famous economist once wrote about the "creative destruction" of capitalism. Strong economies and productivity growth do NOT guarantee that businesses will prosper; on the contrary, they guarantee that some businesses will FAIL. but the ones that fail are the inefficient, less productive businesses and the ones that prosper are the less inefficient, more competitive businesses. in an economy where lots of businesses DON'T fail, you end up with a bunch of companies run like the US Postal Service.
Yeah, I understand the belly-up theory, and everything okay. Of course now we're in Bush's "world of marketing", and I've noticed he runs things a little differently then other Presidents do.
Where's a guy like Brad DeLong when you need him?
Teaching at some university? I guess I should sign-up and pay the tuition, and who knows I might learn something.
"...and who knows I might learn something."
It is a huge mistake to believe that only the Marxists oppose free market policies. Capitalism is not a philosophy of the conservative reactionary. It is in fact very liberal and a destroyer of the comfortable status quo. T.S. Eliot, Russell Kirk, Erza Pound, and countless other men of the Right were sometimes even outright hostile towards the unsettling effects of Capitalism. Please note that Pat Buchanan intensely dislikes neo-Conservatives like myself.
J-Lo, I agree that there are potential, serious, productivity gains out there being had, or waiting to be had. Where I first talked about "transactions costs" and "production technology", I'm acknowledging those very issues. The issue is still about the EMPIRICAL significance of these. Thw wheat is still being sorted out from the chaff here.
(Nice dress by the way. If productivity in dress-wearing terms is measured in percentage-of-skin-shown, your productivity is definitely improving.)
For Cheryl T and for Anarchus: productivity growth in aggregate doesn't address individual businesses, only what it's POSSIBLE to produce with given resources within an economy. It's also a double-edged sword at the macro/business-cycle level; for reasons why, check out Paul Krugman at:
I think there's been more than enough intoning of the Mantra of Creative Destruction in the comments at this blog lately.
PS: Gerry Garvey. Don't be a stranger here. Come back and heckle some more! :-)
I never know where I am in this discussion.
Is the production unemployment or IT?
But what does this mean.
Technology is dead. It needs a whole new definition.
Intellectual patents give us people like Michael Jackson. The Chinese have already adopted Linux. Bill Gates offered India 100 million to fight AIDS and they adopted Linux. Did you see what the Canadians did with the Harvard mouse?
Steal This Book.
Is there an attempt by the Islamic community to return to the gold standard?
Did you see what the Canadians did with the Harvard mouse?
Before anyone here rushes to start fulminating about those backward, parasitic Canadians again (yes, DT, that means you)... I should point out that the Canadian Supreme Court decision only said that the current wording of our patent legislation was insufficiently specific, that this was a complex issue that had not been thought of when the patent law was written, and that new legislation was required to decide exactly where to draw the line on what should and should not be patentable in this area.
In other words, they said, "Over to you, politicians. This goes beyond the scope of what can be decided by the courts."
What happens next is a public debate on the subject. Perhaps there will be a Royal Commission (a peculiarly useful Canadian institution, remind me to explain it sometime). A new bill will be introduced by the government, and it will pass.
RE: For Cheryl T and for Anarchus: productivity growth in aggregate doesn't address individual businesses, only what it's POSSIBLE to produce with given resources within an economy. It's also a double-edged sword at the macro/business-cycle level; for reasons why, check out Paul Krugman at:
. . . . . there's nothing I wrote that conflicts with anything in the pk archive, and I didn't claim that there was a direct linkage between specific companies such as IBM and aggregate economic data. HOWEVER, IBM's sales in dollars ARE a component of GDP just as the people who work for IBM are a component of US employment.
Cheryl's comment basically wondered: "with so many companies going bankrupt and other companies laying people off, is it possible that the productivity data is false?" And the answer is (just as the PK archive points out): No, in fact strong productivity growth means that the businesses in the economy are producing more per person employed, which at the margin may have negative implications for employment.
There is one claim PK makes that is in fact incorrect . . . . . . while most people may think that falling unemployment as a percentage of the labor force is the measure that tells whether or not the economy is performing well, in fact it is not. The only way that any society gets wealthier over time is when output PER CAPITA increases.
> There is one claim PK makes that is in fact incorrect . . . . . . while most people may think that falling unemployment as a percentage of the labor force is the measure that tells whether or not the economy is performing well, in fact it is not. The only way that any society gets wealthier over time is when output PER CAPITA increases.
Performing well for whom? That's the question.
Well, they say a recession is when a neighbor loses his job and a depression is when you lose your job . . . . . . . .
But the answer is, the populace as a whole. Unemployment (at least as they measure it) is at about 6% right now - for the people who want work that can't get it, it's miserable and wretched, no question, but most people can and do have jobs.
If economic policies put in place today generate an extra 1% of productivity growth over the next decade, EVERYBODY ends up better off by a lot, even the least successful of the populace - because the better off a society is (and they only get better off by producing more per capita) the more opportunities there are for everybody.
It's mostly a semantic question, Anarchus.
I can't think of a case where an economy would be performing "badly," but unemployment keeps dropping. By contrast, our current situation is an example of an economy performing "badly" where productivity keeps going up. Not that we should try for less productivity, but it's a pretty useless "good thing" when unemployment keeps rising.
Most importantly, Krugman was pointing out that using "rising productivity growth" as evidence the economy is doing just fine, when GDP and unemployment aren't doing too hot, is a bit of dishonest PR for the government.
Jason: Thanks for the informative comments.
At the moment I think productivity and GDP are doing pretty well. If you take the average of the last 4 quarters' real GDP growth, it's running around +3% to +4%. That's not bad at all.
Employment is not doing well, but employment is a lagging indicator, and current job trends do not fully reflect the improving economy.
The problem with the GDP, productivity and employment math is that taken to its logical extreme, one ends up with the Luddite conclusion that productivity growth is bad because it causes unemployment . . . . . so if we did away with all the bulldozers and back-end loaders and built roads by hand, we'd be so much better off because unemployment would decline sharply.
One point to make about the "productivity gap" is that the economy must grow even faster to increase employment. Currently, GDP needs to be at least 3.5%-4% to add jobs. Also, the "productivity gap" is a deflationary pressure on the economy, is it not? I.e. produce more, more cheaply, yet wages stay the same and employment growth is slower.
what do you understand by productivity and what is it's significance?