The behavioral economics of data consumption

By Felix Salmon
December 10, 2009
you use less of it. But I always thought that was because the more electricity you use, the more you pay. In fact, it might just be a function of the fact that it's being measured and reported back. I was puzzled by this story in today's NYT:

" data-share-img="" data-share="twitter,facebook,linkedin,reddit,google" data-share-count="true">

I know that when you know how much electricity you’re using, you use less of it. But I always thought that was because the more electricity you use, the more you pay. In fact, it might just be a function of the fact that it’s being measured and reported back. I was puzzled by this story in today’s NYT:

Mr. de la Vega cited the heaviest data users, saying that 40 percent of AT&T’s data traffic came from just 3 percent of its smartphone customers.

But he emphasized that the company would first focus on educating consumers about their data consumption in the hope that doing so would encourage them to cut back, even though they are paying for unlimited data use.

Then I saw the WSJ explain:

Many customers don’t know how much bandwidth they’re consuming, Mr. de la Vega added. When AT&T conducted a broadband test, customers often reduced their data use.

I’m not entirely clear on what the mechanism is here: why would you reduce your consumption of something you’re getting an unlimited amount of for free, just because you’re told how much you’re using? The only thing I can think of is that some people are turning off wifi to preserve battery life, which means that they use the 3G network more, and when they find out how much they’re using 3G, they realize that they’d be better off turning wifi back on. But it doesn’t really matter why it works, if it works.


We welcome comments that advance the story through relevant opinion, anecdotes, links and data. If you see a comment that you believe is irrelevant or inappropriate, you can flag it to our editors by using the report abuse links. Views expressed in the comments do not represent those of Reuters. For more information on our comment policy, see

One possible reason is that the person may not have realized that they downloaded an app/virus/bug that may inadvertently be using a large chunk of their bandwidth. That consumption would degrade performance for the user and fixing the problem would improve their experience with the device.

The same could probably be said for electricity. If you find out that you’re using a huge amount of electricity that is outside the norm for a dwelling of your size, then you may be prompted to try to correct the problem (bad wiring, inefficient appliance, etc.) that you may not have even realized you had.

Posted by spectre855 | Report as abusive

My guess is that people are also simply shamed into reducing their usage by the appearance of excess. Probably a name for that psychological phenomenon, but I don’t know what it is.

Posted by runsvoldmd | Report as abusive

I suspect tethering is a factor here. nnen/techwatch/simple-secret-iphone-teth ering-fix AT&T’s previous comments with respect to the top 3% indicate these aren’t simply people who watch a lot of YouTube. These are another class of user. Assuming these are users who have made their iPhone into a primary internet connection via 3G tethering to their laptops, the behavioral economics make perfect sense. A nudge, indicating that AT&T is wise to their off-label use of the network, might be all it takes to get most of these tethered users to pull back their data usage to more normal levels, in hopes of maintaining the tolerance (,2817,23 49349,00.asp) henceforth shown.

Posted by magicc | Report as abusive

There was an experiment in England, I believe, in which energy consumers were told, on their bills, how much energy their average neighbors used, and a lot of consumers moved toward the mean — low consumers started using more, while heavy consumers cut back. When they started putting smiley faces (no kidding) next to the comparison for households using less than the average, their use went down.

Humans are social animals, and social proof — taking cues from others about appropriate behavior, and desiring to conform — is often a strong influence.

Posted by dWj | Report as abusive

I agree with spectre855 above. Felix, if my memory has not become total Swiss cheese, you have mentioned on this blog that that NYC apartment of yours is somewhere south of East 14th, which means it could easily have steam heat and your water bills (same water goes to make heat as to deliver hot water) may be included in your rent or whatever payments (I don’t want to presume, it has been a while since I lived in the Alphabet or New York at all. Plus since Tompkins Square Park has reportedly turned into a dog run, maybe the projects at Avenue D and Houston are now sustainable condos with green roofs and a Waldorf day care on the ground floor).

However, in most places in the US, single-family homes usually receive metered water bills every month or two. If you see your bill jump ridiculously, you figure the toilet(s) is (are) leaking, and you either get the plumber in or try to fix this yourself by sticking your arm in the tank. If this fails, you get someone to find the leak, which by now may be in a wall or under the lawn or something. But yeah, definitely, it means I have a leak.

That said, our local alternative weekly publishes an annual list of the top 10 water users in our metro area, and while several always yell, “Oh no, I must have a leak!” there is always also some prince who says, “Well, we have an awful lot of plants, and there is a dry season of three months here west of the Cascades. Besides, we pay for it.”

Posted by SelenesMom | Report as abusive

Way OT — Felix, the new pic makes you look like the long-lost triplet Proclaimer. Presumably, the one who refused to walk 500 miles.

Posted by SelenesMom | Report as abusive