How Facebook and prostitution distort US GDP numbers


Anyway you figure it, America’s $16 trillion economy is ridiculously large. But as big as it is, it’s probably bigger than government data show. For instance: the EU is making the UK estimate the value of drug deals and prostitution, and then add that amount — some $17 billion — to official output numbers. The US shadow economy, everything from illegal activity to off-the-books work, has been estimated at $2 trillion.

But even that doesn’t catch everything. On my recent Ricochet Money & Politics podcast, I asked AEI economist Stephen Oliner about how technological progress distorts GDP numbers:

One other point that [The Second Machine Age authors Erik Brynjolfsson and Andrew McAfee] make is that they don’t think the current economic statistics do a good job of capturing the rate of innovation of technological progress. Do you have confidence that we have a pretty good feel for the reality of the situation if you look at the productivity data?

I don’t feel entirely confident. I mean, this is one of the things that I’ve been doing research on is whether the economic statistics properly capture the impacts of information technology. And I think there are reasons to be skeptical.

I think the particular thing that Brynjolfsson was talking about is that the things that are available for free on the Internet, you know, setting up a Facebook account, for example, because they aren’t priced, they aren’t part of measured GDP. Even though they have value, people get utility out of those products, but they’re not priced, and therefore they’re not in GDP.

What I’m talking about are whether the statistics that we actually have are distorted in some way. And one example that reflects the research I’m doing now concerns the prices that we measure in the producer price index, which is the U.S. official price index produced by the Bureau of Labor Statistics for semiconductors, particularly the microprocessors that go into computers, laptops, desktops, tablets, et cetera.

The PPI shows that the price declines for those goods, which were extremely rapid throughout almost the entire history that they’ve been produced, have basically come to a halt — that in the last couple of years there have been no price declines to speak of at all, which is very strange and is in conflict with the fact that innovation in that part of the economy is still proceeding at a rapid rate.

And it raises questions about whether the procedures that are being used to measure those prices are appropriate. And I personally think that they’re not, that prices are actually falling more rapidly than the official statistics would show.

I could go on for longer, but I do agree with Erik’s basic point that the measurement framework that we have in the United States does a pretty good job of capturing IT, but there are problems. And I think, overall, it understates how much benefit we’re getting from information technology.

Follow James Pethokoukis on Twitter at @JimPethokoukis, and AEIdeas at @AEIdeas.

4 thoughts on “How Facebook and prostitution distort US GDP numbers

  1. Microprocessor prices have been pretty stable, but prices for video memory, desktop and laptop in-line memory modules, and especially for flash memory, have dropped rapidly.

    Thumb drives (I just purchased a 64 GB USB 3.0 very fast thumb drive for less than thirty dollars) have all but replaced optical media for applications like the transfer of large files. Those transfers also occur in the cloud, suggesting that technical progress is occurring but is not reflected in the statistics.

  2. Right-wingers are nuts (I am in this group) and here is another reason:

    Right-wingers say we are radically underestimating real income and wealth, and thus wildly underestimating real income gains in the middle class. The middle class has it great, stop sniveling. This suggests a huge overcount of inflation.

    But right-wingers are also hysterical about the CPI, and that it is “underestimating” inflation.

    I realize the right-wing is not a monolith…but even within the AEI website, you can find both sentiments….

  3. Seems to me that Facebook economic activity is counted. The reason the account is free is because it is paid for with advertisement. If you assessed some additional economic charge because you can use the account “for free” you would be double counting.

    An older example should advertising based TV be counted twice in the economic numbers because I can watch it for free – or should we just count the advertising revenue?

    • FM radio would be a better example. We get value but pay nothing. Should Old Tech FM radio be counted twice in the economy? Of course we do pay by listen to adds just as we are inundated with adds on Facebook.

Leave a Reply

Your email address will not be published. Required fields are marked *

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>