Arnold Kling  

David Brooks, Tyler Cowen, and PSST

PRINT
Henderson Review of Bryson's "... Why Do So Many GMU Economists ...

David Brooks writes,


In other words, as Cowen makes clear, many of this era's technological breakthroughs produce enormous happiness gains, but surprisingly little additional economic activity.

This is Tyler Cowen's Internet story. Lots of value, but not much economic activity.

It feels wrong to me. If I use the Internet, then other people are voluntarily doing something for me, so it meets my criteria for economic activity. I think that what is going on is that I am getting a lot of consumers' surplus relative to what it costs. That is not stagnation.

Imagine that a lot of people became hooked on virtual reality and cut back on purchases of traditional goods and services. To me, that will change the patterns of specialization and trade. Presumably, people will produce less traditional goods and services and instead produce goods and services that are more complementary to virtual reality. Everyone may choose to take more leisure, because virtual reality generates much more happiness per unit of labor. But the end result (and it may take decades for people to adjust) is not mass unemployment, and it is certainly not stagnation. You may find such a society repugnant to contemplate. But, properly measured, its productivity will be far higher than its predecessor.


Comments and Sharing





COMMENTS (16 to date)
D. F. Linton writes:

His booklet is quite good and some of the negative reviews seem poorly founded, but on the issue of the internet Cowen does seem to confound the idea of economic activity unmeasurable by Government statisticians with activity without economic value. Perhaps it serves his theme better, but as you say, the internet everyday produces a huge bounty of consumer surplus.

Matt C. writes:

I too have a problem with this. I buy a bunch more stuff online than I would have otherwise. For example I can buy something online and pick it up at Best Buy. This saves me time, I am economizing on my time. Megan McArdle has a list of items for Christmas presents, I used it to help me make a decision on one of my wife's presents several years ago. Don't even get me started on Amazon.com and how many jobs have they created? What about Zappos.com? What about ebags.com? The list is nearly endless. Ask FedEx or UPS if their business would have been the same without the internet.

I have a friend who probably has more music than Tyler could even fathom(yes, I know Tyler's a music junky). My friend uses the internet to get items that he couldn't get when he was younger. OR he can now get items that he might not have known about without the internet making it easier to know that there is a different album for one of his favorite bands in a different country. The internet makes things easier and more efficient.

Without the internet we wouldn't have things like the iphone or iPad or droid phones or droid tablets. Yep...not much economic activity there.

But this information is just anecdotal and comes from a very small set of observations, so maybe it's just me.

DW writes:

Agree with DFL, I think if you incorporated the measurement problem into this post, you'd probably find yourself closer to TC.

The Unbeliever writes:

But won't the labor-free nature of marginal unit purchases of virtual goods guarantee lowered overall employment?

You don't even have to posit virtual reality; take iTunes as a real world example.

If I go to a physical store and buy a complex jigsaw puzzle for $10, I've drained some inventory which will need to be replaced. It cost some labor to create the puzzle, package it, ship it, stock it on store shelves, sell it to me at a checkout counter, etc. The additional employment generated by my purchase is equal to the marginal amount of labor needed to create and sell one puzzle (firms anticipating future demand in past periods) plus the cost of replacing the inventory (the usual replenishment cycle).

If I forgo the jigsaw puzzle and go to the iTunes store instead, buy a Bob Dylan album for $10 and download it, I have engaged in the same dollar amount of economic activity. But the good carries exactly 0 labor requirement for each marginal purchase; by copying existing data files, the product I consume was "created" by software at the precise moment I clicked the Buy button. A million consumers could buy the album without any additional labor needed.

(Granted, there took a small amount of labor to put an album up on iTunes in the first place. But this becomes a fixed cost of initially creating a product, and it scales to infinity instead of requiring another fixed cost investment at intervals as sales grow.)

Multiply this by several million consumers, and you can see where virtual goods replacing physical can start to decrease the overall employment level needed to support the same level of consumption. If ten million consumers decide to download a $10 album instead of buying a good or seeing a live show, that is $100 million transferred to Apple (who has no need to hire more people) and $100 million less in services or physical goods that need to be produced.

The same holds true for software, online streaming videos, smart phone apps, etc. There are now so many ways for consumers to purchase leisure without requiring labor to provide it, that it must eventually have a noticeable affect on aggregate employment.

Doc Merlin writes:

"In other words, as Cowen makes clear, many of this era's technological breakthroughs produce enormous happiness gains, but surprisingly little additional economic activity. "

Thats not a very good sentence. The end should read, "lots of value, but not much GDP." This should be an inditement of GDP and the formalisms it represents not of the internet!

Oh that sun! With its free light! Oh what economic uselessness! /SARC

I too would count patterns of sustainable specialization and trade as "innovations". If you do so then I'd find it hard to sustain the claim that the last 30 years haven't seen a huge increase in the rate of such innovations.

J Mann writes:

Isn't there some economics hypothetical chestnut about magically increasing everyone's wealth? (E.g., you wake up tomorrow to discover that someone has discovered a free source of energy).

It's hard to imagine that suddenly making encyclopedias and porn free and accessible to nearly all counts as a REDUCTION in economic activity, even if it counts as a reduction in measured economic activity.

I agree with Doc Merlin that this is more an indictment of measured GDP than a reduction in economic activity. I'm getting more value out of my smartphone, cable internet, aircard, etc., but I'm paying less for them.

Finch writes:

I think Tyler's response would be something like: "Sure, but these distortions in GDP and CPI should have existed in earlier periods as well. Why are they stronger and more important now?" I think he would express doubt that they actually are stronger now.

I would disagree, and I think most people would, but it's hard to convincingly argue why. It sure seems like modern innovations are harder to measure than those of the 1950s or 1900s. First of all, there's just a lot more inventing going on right now, so there are more things to measure. The number of people employed in inventing (basically the engineer count) is higher than it has ever been, and those engineers are more capable than they've ever been. And the price and cost of a lot of these inventions seems near zero, which you might expect would throw things off. But these two objections aren't very quantitative. I don't know that they explain the whole effect.

drobviousso writes:

I think there's a bit of confounding of terms here. When some people say "Internet" they mean websites that people visit to read, shop, and consume content. When some other people say "Internet" they mean that big system of interconnected computers that facilitates the other Internet, as well as corporate email, VOIP, wiki's, shared network drives, and the like used by, for example, my firm for our internal infrastructure.

The first produces more happiness than economic activity, the second is the other way around.

Milton Recht writes:

A lot of the argument about the effect of the internet and long term measures of real GDP and productivity depends on accurate measures of inflation. The internet allows for easy price comparisons and I find I am often buying things on the internet instead of locally because the total cost (including shipping is lower). The price of the goods has not declined but I can avail myself of a better price than from a local merchant and use the leftover funds to make an additional purchase from another merchant. Total nominal and real GDP is constant (I still spent my X dollars) but my welfare has improved.

Also, the further back in time one goes with any data series that depends on inflation adjustments, the more uncertainty there is as to its accuracy.

Over time, inflation measuring and adjustment methodologies change as well as data gathering methods. Long term inflation adjusted data comparisons are fraught with many problems. These types of comparisons work best over short periods.

Evan writes:

The Internet is, partially at least, a gift economy. For this reason it's hard to measure economic activity because many people produce happiness-increasing goods in exchange for status and recognition, rather than for cash. Someone who posts old unavailable TV shows on Youtube, contributes to an open source project, or adds to an encyclopedia gets a feeling of joy when someone comment that they enjoy their work. Essentially they're cutting out the middle-man and getting happiness directly in exchange for generating it.

Chris T writes:

The web has reduced the cost of information distribution to virtually zero. This was a reason many 'dot-coms' failed during tech bust - their business model was predicated on information having a cost.

One can see the havoc this is creating on other information industries, such as newspapers.

Drobviousso brings up a good point; make sure your clear what precisely you're referring to when you talk about the internet. The web runs on the internet, but is only one application of the internet.

Lord writes:

One has to consider costs as well as benefits. For all that it has created, it has destroyed much too. It may have even destroyed more than it has created hitherto, thus the search for new patterns and ways of monetization. This is not to say we won't be better off eventually but for diffuse but widespread gains we have to endure concentrated and intense losses. The difficulty of the change suggests losses have outweighed gains to date.

Ryan writes:

I too agree with Doc. The Internet, in it's omnipresent form, has had a substantial positive impact on leisure -- this includes the time "The Unbeliever" saves by purchasing online. Leisure is not a factor of GDP thus likely doesn't constitute "economic activity" by the book's definition.

Evan writes:
the Internet, in it's omnipresent form, has had a substantial positive impact on leisure -- this includes the time "The Unbeliever" saves by purchasing online.

It also increases the demand for leisure, however. Since it is now cheaper than ever before to obtain many goods, people will now forgo the money they previously needed to purchase them in order to obtain the leisure they need to properly enjoy them.

As a personal example, I am an old film buff. In a previous decade I might have worked fantastically hard, and pursued high-stress, high-paying jobs in order to fulfill my hobby. It would be the only way to afford to buy old movies, canvas stores for them, attend special screenings, and purchase viewing equipment.

Now I have Amazon, Youtube, and Hulu to aid me in my quests for obscure films. Hence my main problem is not getting enough money to obtain them, but rather getting enough time to watch them. Hence I went for a more leisurely job.

Steve Roth writes:

Arnold, not really about this post, but would be interested to hear your comments on this:

http://krugman.blogs.nytimes.com/2011/02/13/whos-unemployed/

Comments for this entry have been closed
Return to top