What is the economic value of the internet?

There is a new and very interesting study out from McKinsey on this topic.  If you get past the press release, however, and give it a closer read it is consistent with stagnation hypotheses, contrary to some claims.  The study shows a few things:

1. Most of the economic benefits of the internet are in fact captured in current economic statistics, which I’ve already argued do not look so fabulous.  The point is not to blame the internet, as without it things would have been worse.  The point is that the internet gains, in absolute terms, haven’t been large enough to produce a rosy picture overall.

2. The direct and indirect economic effects of the internet account for 3.8% of U.S. gdp, as currently measured (p.15).  That’s less than many people think and that value is already incorporated in the current gdp measure.  The good news — and it is good news — is that there is lots of room for future growth from internet impact.  We’ve yet to really organize our economy around the internet, as we someday will, and then the gains will be enormous.  In the meantime we are waiting.

3. What about the unpriced consumer surplus gains from the internet?  The study considers that too:

In general, this surplus is generated from the exceptional value users place on Internet services such as e-mail, social networks, search facilities, and online reservation services, among many others. This value far outweighs the costs, both actual costs such as access and subscription fees and annoyances such as spam, excessive advertising, and the need to disclose personal data for some services. In the United States, for example, research conducted with the Interactive Advertising Board found that consumers placed a value of almost €61 billion on the services they got from the Internet, while they would pay about €15 billion to get rid of the annoyances, suggesting a net consumer surplus of about €46 billion.

A nice gain, but in an economy with a $14 trillion gdp it’s not that big a deal.  It’s also less than the previous Goolsbee and Klenow estimate of about two percent of gdp, for the consumer surplus from internet use.  That is not nearly enough to refute the view that median income growth is much slower in recent times, and that’s without adjusting for the problematic status of expenditures on health care and K-12 education.  Arnold Kling and Bryan Caplan stress this point — the unpriced benefits of internet use — but I don’t see them offering a better number than this rather paltry calculation, now confirmed as low, at least relative to the claims of internet boosters, by two differing sources.

In fairness, I should note that I cannot trace the study cited above.  It may be wrong.  The Goolsbee and Klenow paper may be wrong, and it is somewhat out of date, from 2006.  I think a consumer surplus estimate of three or four percent is entirely plausible for 2011.  We’re still left with relative stagnation.  After all, penicillin had some consumer surplus too.


We are still in the middle innings of The Great Recalculation.

Sorry the game has yet to start. We don't know yet Internet's full potential and how to use it well. Any economic evaluation of the benefits and costs over the next decades or centuries cannot ignore how little we know about Internet, in particular about when 99,90% of the world population (the relevant set of people to evaluate Internet) will be able to use it effectively. Yes, I include myself in that 99,90%.
In addition, Tyler should have started by discussing the methodology presented in the Report's appendix and try to make sense of it given the complexity of Internet as a resource. If in 1900 someone had asked you a valuation of U.S. railroad system, how would you have done it? What if today --read last Tuesday's NYT editorial-- you were asked to put a value on Chile's Patagonia? To discuss the conclusions, first we have to agree on a methodology for its economic evaluation. Yesterday I met a colleague with whom we made our first economic evaluation of an infrastructure project (in 1962, see http://www.cabracorral.com/cabra.php ), we talked about how much methodology has changed since then and how our experience over decades of professional work was not enough to evaluate some resources like Chile's Patagonia.

I mainly view the internet as a cost and waste reducer. The problem is that in the short term, one person's cost/waste is another person's profit, upon which they made a lot of bets and debts.

Talking out of my asset class here obviously, but I'd start the methodology, as Tyler has started, by attempting to explain the lackluster economy despite the enormous potential. That's partly how I arrive at the internet as mainly undermining economic moats and creating new ones.

All humans' inventions have reduced the cost doing old activities and created opportunities for new ones. Many inventions before and after Internet have been related to a small, well-defined set of old activities and a few new ones closely related to the old ones. If you want to evaluate one of those inventions the past provides a lot of relevant information.
When we evaluated Cabra Corral in 1962, our principals agreed that the river basin and the surrounding areas had just one alternative use --agriculture with traditional technology-- and that the irrigation project had to be only for expanding agriculture (including new outputs and new technology). I understand the decision to build it was based just on this expansion. Only later other activities were considered and justified largely by the sunk cost of the original project. Today we debate about the value of natural resources like Chile's Patagonia and all sorts of people argue that they would be willing to pay something for maintaining the resources free of any human activity (fundamentalists even argue that we should not visit them) or for undertaking activities that will have a low environmental cost (fundamentalists always know a better way to do things, in particular to produce energy). Not surprisingly, the evaluation of particular uses of natural resources has become a major challenge.
Internet is a special type of resource because as an infrastructure project --contrary to what happened with new means of transporting people and goods in the 19th Century-- from the very beginning it was expected to be a platform for a large number of new activities in relation to the old ones that would benefit from it. I have not studied Internet activities to know enough about its potential in terms of undertaking new activities and reducing costs of old ones, but I believe that almost all the world population will benefit from Internet by 2050 (the cost of accessing and taking advantage of Internet has declined sharply in the last 20 years and will continue to decline to levels that will make it affordable to most of the world population).
Even if the economy were as lackluster as you think it is and the prospects were as gloomy as Tyler's TGS implies (I disagree because I'm interested in the world economy not just in the U.S. economy), I'm inclined to think that Internet activities (old ones and new ones) will contribute to tear down many of the remaining Chinese walls that still limit the integration of the world market economy --within large countries as well as among all countries. Indeed I'm very much in favor of economic integration.

"Even if the economy were as lackluster as you think it is"

I don't think it's as lackluster as you think I think it is ;)

How do they calculate what value consumers put on something? Is it based on how much they might pay for it? That will never capture the true surplus. The values cited above work out to about $266 per person, which I assume is per year.

I will pay about $500 a year for internet access because I know I can get it for that. But if I imagine what my life would be like without Internet, I would conservatively value it at around $20,000 per year. Even if you only count money saved on gas, stamps, maps, and online coupons, the value is well over $266 per year. But the internet does so much more than that.

Recently the subject of culling was discussed on this blog. Simply as a culling tool, the internet provides immeasurable value as a method to select books, magazines, movies, TV, theatre, music, vacation destinations and restaraunts. Its full of great resources to learn about history, science, and current events. Think of all the hobbies and DIY projects that would be far more difficult without the internet. Once they figure out a way to put sex on the internet, it will provide the full range of the human experience.

How would this percentage of GDP compare to that of older, existing technologies 15 years out from their launch? For example, what percentage of GDP were railroads, airplanes, desktop computers, etc. at an equivalent stage of maturation?

First of all, we need a term for the whole constellation of modern devices, technologies, and information & entertainment services that all work together. Maybe the 'extended internet' (after Dawkins 'extended phenotype')?

Regardless of what we call it, assuming that's what we're talking about when we say 'the internet', I'd much sooner give up indoor plumbing and central heat. I'd live in a house a half or a third the size of my current one. Better life in a log cabin with a wood-stove, a well and an outhouse than one cut off completely from 'the internet'.

And there's good reason to believe this is NOT just because I'm an advanced-degreed, Marginal Revolution-reading informavore. Recall the recent article about the bottom billion:

We asked Oucha Mbarbk what he would do if he had more money. He said he would buy more food. Then we asked him what he would do if he had even more money. He said he would buy better-tasting food. We were starting to feel very bad for him and his family, when we noticed the TV and other high-tech gadgets. Why had he bought all these things if he felt the family did not have enough to eat? He laughed, and said, "Oh, but television is more important than food!"

Damn right -- and satellite TV is just a small subset of 'the internet'. When somebody on the margins will skimp on food for himself and his family just a small slice of 'the internet', I don't think 'not a big deal' is really a supportable conclusion.

This seems like a serious economic and development issue, BTW. If revealed preference shows people value 'the internet' over even fully adequate nutrition, should governments and development agencies redirect resources and effort into expanding access? And should it matter if people want 'the internet' for productive, wholesome reasons that development agencies would approve -- or if they want it mostly for facebook+porn+bollywood/telenovelas+LOL cats?

I like where you're going with the spirit of the thing in terms of "what else the internet helped spawn" but I'm not sure that satellite TV (and by extension TV programming) qualifies.

Especially since consumer mobile phones and cable TV predate the consumer internet.

I'm pretty sure that the majority of 'net users (in the US for sure and probably also true worldwide) are watching programming via satellite, coax cable and a box, or DVD.

Exactly right. To say that the internet "adds little value", you have to ignore huge legions of reality starting you in the face. The right measure is how much people would pay to just barely prefer having the internet to the loss of money. And that's a f***ing lot.

If I find a product I never would have known about but for the internet, which adds tremendous value to my life, and I redirect $X of my spending to it, how does that show up in GDP? It doesn't. It might even shrink it. Yet only a Tyler Cowen-type economist would conclude that "Gee golly, this doesn't show up in any economic measure devised in the '30s, and you know, it would blow a hole in my trollish thesis, so I guess it must not add any value."


Sorry, as Tyler notes, they do address, this, giving him a chance at another error:

A nice gain, but in an economy with a $14 trillion gdp [61 billion Euros] not that big a deal

Yeah, it sure isn't, when you compare apples to oranges. You should compare:

- the consumer surplus derived from the internet to the consumer surplus in the rest of the economy,

OR, less helpfully,

- the expenditures in the internet sector to the expenditures elsewhere


- consumer surplus (utility NET of expenditure) in one sector to total expenditures (GDP) elsewhere.

Why not derive the consumer surplus from 90% of the financial sector? How do you think that compares? "Oh no! My life would be so empty if I could have all these ultra-complex mortgage "products" to choose from! Save me!"

Also, does this account for consumer surplus in a lot of things the internet makes possible, but is completely hidden from us. What's the consumer surplus of being able to swipe your credit card at a supermarket rather than being forced to use cash or checks because they couldn't swipe at the POS to confirm credit card approval? What about the value of email to keep in touch with family or to use at work?

Here is the IAB study:
There are some things I do not understand in the paper. For example, it is unclear whether they are lumping the U.S. and Europe together in some of their numbers, or treating them separately. The U.S. and Europe have different privacy and security laws, which have different implications for the applications provided, and the study does not address that. Still, as even a rough estimate of the consumer surplus, it suggests that it is large relative to the costs, but small relative to the overall economy.

Also, the Census Bureau estimates that Business-to-Consumer Shipments in 2008 were $288 billion, to give you an idea of the total size of the internet economy in priced final goods.

Oops, here's the Census Bureau link:

Oops...posted as a reply above by accident

How would this percentage of GDP compare to that of older, existing technologies 15 years out from their launch? For example, what percentage of GDP were railroads, airplanes, desktop computers, etc. at an equivalent stage of maturation?

This is a very good question. It is not that the internet hasn't brought about large changes, but that the losses have nearly overwhelmed the gains. Once these losses are mostly behind us we may experience sizable gains. Many of these went through large boom bust cycles as investments were made that didn't pan out only to be redoubled. Most of the gains were indirect and only realized over the course of time as they became integrated into the full economy. For everyone that says it has changed their lives the question must be changed it from what and what was that worth before it was destroyed. How much has the media and publishing industry been destroyed? How much has it enabled the destruction of local economies and offshoring? We may be, will be, better off in the end, but the end is quite a ways off.

I think there's a question of utility too -- e.g., what is the economic value of being able to socialize on Facebook?

This kind of question makes me grind my teeth and think the Austrian distrust of econometrics was well-founded.

Google maps (and Earth) alone is worth $266/year.

I have to agree: gains from the Internet are low because it is derutilized relative to its potential. I'd say that should be obvious, but, if it were, it'd be better utilized since we'd all have a good idea of what to do with it.

My life would be dramatically different without the internet. Its hard to even imagine.

So, people who think it worthwhile to drop comments at a website think the internet is really, really important. I wonder about a negative correlation between the internet being a highly valuable thing in a given person's life and the value of that person's life in the nation's economy.

If I had to, I would pay $300/month for internet.

How much if 'the internet' included all interrelated electronic technologies and services that were unavailable in, say, 1973 (personal computers, video game systems, mobile phones, GPS systems, cable TV, DVRs, digital cameras, MP3 players, ebooks, etc -- as well as all the commercial, information, and entertainment services enabled by those devices)? Still just $300/month?

Exactly. How can we evaluate the value of something that is so embeeded in life that people don't even notice anymore? Do you think people remember all internet devices when asked how much they would pay for "the internet"?

You can get REAL women for that kind of money.

My goodness. You're French.

Here's another revealed preference data point:


I spend a lot of time with homeschool families. Most purchase significant percentages of their educational services via the Internet, and most deliver educational services to others via the Internet. They gather for various classes and club activities as well.

Most people I visit have computer problems and the wrong connections for their HD TVs, which are often set to the wrong formats. Huge opportunities exist for young people to offer tech services to older people. But most young people are not legally able to provide these services from their schools or through firms.

I spend a lot of time at a health center/retirement center. These facilities could but don't use the Internet to allow elderly residents to provide services or to receive them. They usually now have wifi but don't allow streaming due to low bandwidth. Old folks mostly have TVs, not tablets.

Who knows what educational services the young and the old would provide or consume if they were not so much locked into institutions unfriendly to value-creation and exchange.

Public libraries sometimes fill this role for older persons by helping them set up email accounts, learn rudimentary web searching, etc.
Kids get it in school.

Why do we have to worry so much about valuations of the internet's worth? There's hardly a large money subsidy (in the larger scheme of the budget) from the government to internet usage is there?

If McKinsey tells you that eating steaks that you love is grossly overhyped will you stop eating them or mistrust the study? Sometimes I trust my intuition over unlikely conclusions ( especially over fluffy McKinsey studies!)

PS. Since when did it become cool to quote McKinsey as a reference in academic circles?

As a fan of Kurzweil, I think we're stuck in the doldrums between the Internet and major forthcoming advances in genetics, robotics, and nano.

A couple more cycles of Moore's Law and we should be able to model biological behavior such as protein folding with considerably better accuracy,

Something else to keep an eye on is fusion. The big money is going to interesting but impractical science projects like NIF, ITER, JET etc which have no plausible path to being economically competitive with fission (and we have thousands of years' worth of fissionables), but there are a half-dozen or so alt-techs (Polywell, FRCs, DPFs, Rossi) that might pan out in the next 5-10 years and for various reasons could be cheap enough to push energy prices way down.

I didn't get the point about penicillin.

GDP is a terrible measure of economic value, especially for a technology that saves money.

Consider all the services that have been automated thanks to the net: all newspapers that don't need to be printed; all closed call centers, all music, movies and books people (illegally) download for free; all travel miles saved etc. None of which show up in GDP figures.

This is an economics website: why not have people bid for the value to them of the internet.

If there were a world without the internet, how much would you be willing to pay? How much would the advertisers, merchants, vendors be willing to pay?

There is a difference between value and cost. GDP numbers represent costs, because they can be counted, they do not measure value, just as a glass of water in the desert may only cost a cent but be very valuable to a thirsty traveler.

Didn't the internet, at least in some industries, reduce the GDP produced. Apparently, that would be considered a blow against the internet, which is crazy. In other words never confuse accounting identities for actual consumer surplus.

The problem with the Internet is the view that it created the impression, to quote a song lyric, "money for nothing" was possible. The revolution of the 90s was fueled my massive investment in both intellectual capital, but also built capital, with consumers spending vastly larger amounts on computer technology to create the power of the Internet.

While most people claim the technology boom lowered costs when in fact costs increased. The misunderstanding comes from the early adopters of technology seeing their costs plummet as the costs to billions of people exploded - how can you claim someone who had no computer and no Internet has saved money by buying a $500 computer - $500 more than they had paid before for computers, and added $40 a month to their spending to pay for Internet access, $40 more per month than they ever paid before.

And access has been greatly expanded by the use of cell phones with texting that provide Internet access in restricted form for a huge increase in monthly spending, that might only be $20 a month, but that is $20 more than the $0 a month they paid before.

The Internet turned computer tech into something that could attract the masses, and the competition forward priced the goods to explode the size of the customer base to greatly increase spending on technology.

The mistaken impression that technology has reduced costs because early adopters get much more for less creates the expectation that the greatest needs of the planet will be solved by cutting the spending on those needs. Thus the solution to food deficits is cheaper food, and the solution to energy is cheaper energy, and people measure cheaper as spending less than today on food and energy. Three billion people are certainly going to need to spend more for electricity because they don't pay much if anything for electricity, and those three billion people aren't going to be buying nuclear power electricity - most pay for solar power capital that produces electricity that replaces kerosene or darkness or wood fires.

Those who are spending vastly more on electricity from solar panels are the same people spending vastly more for technology to gain Internet access via cell phones, and both are the fruit of massive government spending, with the high profile spending in California, over decades. The difference between the Internet and electricity in the US is the US government spent billions decades ago to give everyone access to electricity so today everyone uses government directed capital for unsustainable electricity over the past century as the price basis for any future sustainable electricity capital investment. For the greatest users of technology like Internet, they see the costs falling sharply while the costs for the nation rose dramatically, so they are demanding that any new technology must cost less than the old unsustainable technology.

But when the world demand for oil increases as a billion new auto drivers increase their spending drastically, the market signals it being unsustainable with higher gas prices in the US. Replacing gasoline with electricity is seen as a cost hike, and the desperate solution is to defy nature by drill baby drill to increase the supply of oil in the ground to reduce gas prices. But drill baby drill doesn't reverse the rapid decline of oil in the ground. In the early 80s, the US had a spurt of extremely increased drill baby drill, but the US production of petroleum just fell fell fell, and it has been falling since 1970 with a minor up tick thanks to government directives that dictated an oil pipeline to open up Alaska during the Nixon and Carter administrations. Other oil fields around the world were opened up in the 70s with production coming on line in the 80s, but those oil fields are long past peak and are in decline with demand rising from people who are increasing their spending on transportation by orders of magnitude.

What has kept the price of oil down is huge increases in efficiency in using oil, and replacing oil with other sources of power.

But substitution and increased efficiency require increased spending on capital. What is ironic is those who see the tech boom as a great sign of the free market by reducing spending was fueled by massive spending on capital paid for by massive increases in spending. The median spending on technology in the US is higher today than in 1980, 1990, and I;m pretty sure higher than in 2000.

Yet for all the added spending on technology for a hundred million Americans, what benefit has flowed to the people who are spending all the extra money?

Long post =/= insightful remark.

I think mulp's post is excellent, but it might be more concise to say that the internet DOES add value, by cutting communication costs. Its cheaper and easier to email than to snail mail someone. Pre-internet, I would have maybe seen Tyler's newpaper column and wrote a letter to the editor, or seen a padded article in an academic journal and written him directly (and he would have read and responded to my letter, because the trouble of sending it would be so high that he would not get that many).

This is useful, but its simply not transform the economy stuff. Its not another source of fuel, it doesn't increase agricultural yields, or manufacturing output, or so on. It hasn't even affected much how businesses and governments are organized. It would put the impact at slightly less than the impact of radio, which is a big deal, just not as big a deal as the hype.

The only question worth asking here is: How much you'd need to be paid to stop using Internet.

Add that for each person, and you'll see a very high surplus.

I am not sure that the approaches in the McKinsey report are complete. Th value of Internet can be assessed also with cost-benefit analysis and impact assessments. In one of the comments here it was cited the consumer "willingness to pay" approach and any difference with a baseline scenario of a world without Internet. I believe we should start from imaging a world without Internet to assess its "added value" and consumers surpluses. Then we should also value the cost savings (ex. buying goods and services), efficiency and productivity gains (time and opportunity costs) as well as information (including data use) and competitiveness gains. Spill-over effects on technology industries and amenities are also important to be valued. We should not forget the environmental impact and the benefit of digital. In the report they talk about ecosystem but no mention if environment!

Talking out of my asset class here obviously, but I’d start the methodology, as Tyler has started, by attempting to explain the lackluster economy despite the enormous potential. That’s partly how I arrive at the internet as mainly undermining economic moats and creating new ones.

This shows which they last very much lengthier and thus saving you income which could otherwise are actually utilized to purchase new ones.215

I'd like to point out the internet's contribution towards lowering search costs and thus lowering average prices (and price dispersion). A study of how life insurance prices were affected by price comparison sites (journal of political economy, vol 110) suggested that price comparison sites alone could have increased consumer surplus by $0.2-1 billion per year since they were introduced (in the mid-90's?). The paper was written in 2002 so it doesn't capture the gains since then. I'm not sure if this is included in the €61bn figure of consumer surplus in the original post

Comments for this post are closed