Digital deflation? Not an attractive idea
Commenter Morgan Warstler keeps insisting that I’m dense because I don’t see the glorious digital revolution, which is transforming our lives and dramatically raising living standards. If only I could see that the actual price level is falling, if we took account of all the free goodies provided by the internet. Here’s me and then Morgan’s response:
“Morgan, OK, tell me how fast real consumption is rising, to within 3 decimal points. Show your work.”
Sure!
The invisible real consumption measure is this:
How much less money will you live with today to not have to live yesterday without access to today’s things?
Whatever you attribute to “money illusion” there is a greater gravity pulling the another way.
I don’t deny Money Illusion has a short term weighted value. But I don’t need 10 years to show digital deflation.
Since 2008, measuring real consumption, the quality of American (not easy to benefit 3rd worlders) life has increased at a pace faster than ANY 7 years period pre-Internet in American history.
So here’s my work:
1. Scott Sumner finds the best 7 years in real consumption in American history pre-1994.
2. 2008-today is rising faster than that.
And that’s when he’s being relatively sober, other times he seems like a cross between Gollum and a dominatrix:
“Morgan, There’s is nothing to win because it doesn’t matter what the numbers are; these are just statistics pulled out of thin air. Only NGDP is real. And NGDP growth is slowing.”
SORRY SCOTT
Your precious is a little b*tch compared to mine.
REAL CONSUMPTION measures in Digital > Atomic is the highest moral metric.
YOUR JOB is to argue that your precious at X% best services my precious.
So why am I wasting your time with this? Because even Martin Feldstein, the most unMorgan-like creature in the entire universe, seems to feel the same way. This is John Cochrane responding to Feldstein:
Martin Feldstein has an interesting Op-Ed in the Wall Street Journal, “Why the U.S. Underestimates Growth.”
The basic idea is that inflation may be overstated, because it doesn’t do a good job of handling new products. As a result, real output growth may be a bit stronger than measured. Marty runs through a lot of sensible conclusions.
He doesn’t talk about monetary policy, but that’s interesting too. So what if inflation really is (say) 3% lower than we think it is, and therefore real output growth is 3% larger than it really is?
That would mean we are a lot closer to “normal” of course.
So why am I still skeptical?
1. From 1995 to 2004, productivity and real GDP rose at an unusually rapid rate. The IT cheerleaders told us that this fast productivity growth was the long delayed fruits of the IT revolution. Now we have very slow growth, and the digiterati tell us it’s also caused by the IT revolution, which is generating lots of stuff that doesn’t get picked up in the output data, because it’s free. While I’m impressed by an explanation that’s as flexible as a circus contortionist, I’d prefer something that isn’t consistent with any possible state of the universe. I’m no Popperian, but I like my theories to be at least a little bit falsifiable.
2. Much of this discussion proceeds as if economists have some sort of clear concept in mind when they talk about the price level. We are led to believe that if only we had God-like powers to know everything, we could determine the “true” rate of inflation. Not so, economists have never even figured out what inflation is supposed to measure, in a world where product quality is constantly changing. Is it supposed to be the extra income you’d need today to be just as happy as the average person was in 1950? That’s one definition, but that’s not what we are doing. And if you look at what we are actually doing, it has no theoretical justification anywhere in economic theory.
3. Yes, I agree, fuzzy concepts can be useful. I have no problem with someone saying Britain’s inflation was about 20% in 1980, and 1% today. That’s a fuzzy statement, but it is sort of useful. But for debates about digital deflation to be useful, we need a far better idea of what inflation is supposed to measure. Do most people really feel that a constant nominal salary now means a rising standard of living? If so, I’d expect to see a groundswell of support for cutting the minimum wage, to below $7.25 an hour. After all, a lower nominal income would provide just as good a standard of living, since the poor can now see lots of beautiful pictures of food on the internet. And now they can travel using Uber! Seriously, I think the average person would find the idea of digital deflation to be absurd. On the other hand I’ve never been stopped from believing something just because the average person finds the idea absurd.
4. With the atomic world, there was a sort of logic connecting more stuff with higher living standards. More food, a washing machine, a TV and telephone. But with the digital world “more” begins to seem rather monotonous. I used to really enjoy reading magazines. I recall occasionally leaving a barbershop or doctor’s office and secretly wanting to finish a National Geographic article I had started on while waiting. I only held back because I didn’t want to look “weird” to the receptionist. Information was really enjoyable because it was so scarce. Now there’s a sort of infinite magazine at my fingertips. And it’s all a bit too much. Travelling used to be about discovery, strange new worlds you’d never seen. Now it’s “Yup, Costa Rica looks just like it did on the internet while I was planning the trip.” And what about the negatives, the constant annoyance of your computer freezing up, or losing a long email that you had almost finished typing, and hadn’t saved.
5. Revealed preference? OK, I like that argument as much as the next guy. But who’s to say it’s not like being a heroin addict. Just got to check one more site before I go to bed, the 10 best places to retire, and then suddenly it’s 1 am.
Again, the price level and RGDP don’t matter, only NGDP matters. I suppose it’s interesting to ask whether our living standards are improving, if we had any sort of semi-objective way of doing so. But as far as I can tell we don’t have one, and most economists are oblivious to the fact that we don’t have one. That makes debates over the “true” rate of inflation quite tedious, like people who debate how good a basketball player Bill Russell would be today, without actually having even a clue as to how one would come up with a meaningful answer.
PS. In previous posts I’ve argued that living standards are clearly much higher than in 1973, so I also don’t agree with the opposite extreme, who say real wages aren’t rising at all.
PPS. I’ve never bought a cell phone, but instead have been using old discarded ones. November 29 I get a free iPhone 6 plus. I’ll let you know if it brings me great happiness. I know Morgan will be pleased.
Tags:
16. June 2015 at 19:01
Great post. No complaints from me about this one:)
16. June 2015 at 19:43
Well, that’s just because you personally have not benefitted from digital technology. But, imagine if you were, say, an economist at a small school who was suddenly thrust into celebrity status in a way that could only happen because of the global reach of digital communications that allow content to trump all of the frictions that normally create obstacles to heterodox ideas. Now, if you were THAT guy, the internet would have added immeasurable value to your life, and you would certainly understand Morgan’s point. If that was the case, you’d probably have something like a blog where you could share your thoughts about it and have a conversation with a huge group of like minded people – the sort of thing that a person engaged in ideas would consider priceless. THAT would be something…if you could have been that guy, I mean.
😉
16. June 2015 at 19:44
I’d say your “negatives” of computers are places where they’ve actually really improved to save frustration – Gmail and other modern email program automatically save drafts every few seconds, and old operating systems would crash and freeze way more often than new ones – and require a complete, slow reboot rather than just killing the bad program. Though software updates have gotten much worse.
16. June 2015 at 19:58
Great post indeed (but which post by Mr. Sumner isn’t?). I always get the feeling though that some (actually a lot of) people from the ‘older’ generation don’t appreciate the so-called digital revolution in all its forms. I personally cannot imagine a world without IT anymore (in private life and even more so in my professional life).
Let me give an example: I’m a physician and now every year there’s so much more information in medicine (some of it actually being relevant). So how would you find all this new information without IT? How would I find and prescribe all the new medication? How would I handle all my patients? In my opinion IT changed my profession in a really positive way and so far it get’s better every year.
So from my point of view the digital revolution raised the living standards of physicians and their patients a lot. Not in quantity maybe but in quality.
16. June 2015 at 20:42
Isn’t “digital deflation” irrelevant?
We want inflation high enough that nominal wage cuts can be rare.
We want inflation low enough that loaning or depositing cash can feel safe.
As long as wages are in nominal dollars and people hate wage cuts, no amount of digital deflation would erase the employer’s problem with low NGDP growth.
As long as the bulk of consumer purchases (food! shelter!) aren’t digital, no amount of digital deflation would erase the depositor’s problem with high standard inflation.
I can see how “digital deflation” matters for analyzing “true” productivity growth. It’s sort of like owner’s imputed rent or the unpaid child-care labor of a stay-at-home parent. It’s worth trying to measure it, or at least noting that it’s there.
But this digital deflation doesn’t matter much for analyzing whether we have too much or too little inflation/NGDP growth, does it?
16. June 2015 at 20:45
I think the comparison between the 1990’s computer driven growth and today’s growth is about the application of the technology. In the 1990’s it was about improvements in manufacturing, today it is about improvements in consumer experience. And improvements in consumer experience is really hard to measure, not just because of the difficulty in measuring utility, but because much of consumer experience is not monetized.
Consider a person plugged into a virtual reality machine similar to in the Matrix film. The VR machine is perfect and the person is provided with a life experience that is absolute perfect utility for that person. If the experience ever deviates from perfect utility the machine corrects the experience instantly. The machine was built by free robots which also provide his nutrients and the program is generated for free by an AI. The VR person is not trading with anyone. Measured GDP in dollars for a community of people like this is precisely zero. But utility is off the scale, it’s literally as high as it can be.
So think of a Laffer curve with the x axis being personal utility and the y axis being monetary value of consumption per person (which should average out to GDP per head). You start at the zero axis with zero dollar value consumption per person and very low utility (a sort of malthusian world), as you move up the X axis the consumption $ per head rises to a maximum and then starts to decline as you approach this virtual reality world.
Where we are on this curve I don’t know, I doubt the most advanced societies are beyond the peak of measured consumption yet, but as we approach the peak surely the rate of change per increase in utility has to slow. On a personal basis, many people are already well beyond the peak.
16. June 2015 at 20:46
If we are spending less money for much bigger amount of communication, reading news, books, listening to music, watching movies, etc. and also we save so much time by quickly searching for necessary information through Google or Wikipedia, booking our travels and accommodation online, and a bunch of other things we can do online, in my language, that’s rising real living standards. However, since all that influence cannot be objectively measured, I’m not sure how would it be proper to count regarding inflation and deflation.
16. June 2015 at 20:48
There is a downside to IT in the business world that I hope is studied and quantified someday. In the cost benefit analysis of IT systems, I’ve never seen opportunity costs factored into the analysis. For example, the costs of a project management system will be based upon the licensing of the software, the hardware cost, and the operation costs. That will then be compared to the employment costs of project management specialists. But there is no evaluation of the opportunity cost of having thousands of employees spending a several weeks out of the year to do their own project management work. And those same employees will have to spend time dealing with the requisition systems and change management systems.
Also, the upfront costs of generating information are so low that management does not consider the opportunity costs to have their employees keep up with the mountain of email that gets generated. I typically had to spend at least 10 hours every week to keep up with work email. I can tell you that the company did not get more value from my having to deal with all that information rather than my spending time doing the primary work for which I was being paid.
16. June 2015 at 21:47
Good to see that Morgan has been upgraded from “wacky commenter” to just “commenter.”
16. June 2015 at 21:50
Nothing is free. All that free content we consume has a cost of production, and that cost is picked up in the statistics.
16. June 2015 at 21:55
Wonderful blog post.
I am reading and commenting from a Google Nexus 5, which cost me nothing up-front when I signed a £30/month contract with Vodafone.
Forget 1973, I got my first mobile phone in 1999, and it cost sonething like nothing up-front when I signed a £25/month contract with One2One. I could not access The Money Illusion using that phone.
Tell me life has not improved 🙂
17. June 2015 at 00:23
LOL, great putdown by Kevin Erdmann of Sumner. Sumner rambling post somehow links our own Morgan Warstler (an internet VC-type new concepts salesman, he will not be pleased with Sumner’s post) with Martin Feldstein (Dr. Doom of the 1980s) and manages to mention New Normal in a new way (as a greater rate of growth rather than less). Next post Sumner may mention the provocative Dark Matter and Intl Payments solution by economists Ricardo Hausmann and Federico Sturzenegger?
Much ado about nothing, with lots of spurious correlations and strawmen, a typical ssumner post. Then again Sumner believes the Fed ‘controls’ the economy, so he’s not as non-conspiracy minded and scientific as he claims to be. Show us the study of where it can be shown the Fed controls the economy, even short term? The closest I have found, and it’s not persuasive, is this cite: Lawrence Christiano, Martin Eichenbaum, Charles Evan, “The Effects of Monetary Policy Shocks: Evidence from the Flow of Funds”, Review of Economics and Statistics, February 1006, 78-1, cited in Olivier Blanchard “Macroeconomics”, 2nd edition., p. 96, 5-6, ‘Does the IS-LM Model Actually Capture What Happens in the Economy?’ Shows for data from 1960 to 1990, that a 1% increase in the Fed funds rate appears to show over 4-8 quarters a decrease in sales, decrease in output, decrease in employment, and an increase in unemployment rate up to 6 quarters. However, the “confidence band is only 60% probability (not the usual 95% confidence interval used by most scientific papers). Thus a fraudulent paper.
17. June 2015 at 01:08
I hope it goes without saying, but I didn’t intend my comment as a put down, as Ray claims.
17. June 2015 at 03:32
Shweet!
“While I’m impressed by an explanation that’s as flexible as a circus contortionist, I’d prefer something that isn’t consistent with any possible state of the universe. I’m no Popperian, but I like my theories to be at least a little bit falsifiable.”
Sorry, but no.
We’re all familiar with the brain pan analogy (you are in the Matrix), in which the digital flow of human sensation is invaded by a man in the middle attack… man in middle it looks like this:
Sender computer program receiver
This is Scott’s old economy:
Nature Scott
What we have right now is this:
Natural reality ————–> computer program Scott
Eventually it becomes this:
computer program Scott
Here we see that as Nature is replaced computer program, the constructs of Scott’s economy disappear.
Scott, the first Internet famous economist, mentions odd things like:
1. frozen computers.
2. Lost work
3. But with the digital world “more” begins to seem rather monotonous.
BUT REALLY what stuck out me is this:
“Is it supposed to be the extra income you’d need today to be just as happy as the average person was in 1950? ”
It’s either a typo, or Scott saying deep down, he’s not accepting the core argument of digital deflation:
ALL HUMANS TODAY are better off (and thus economically happier) than in 1950.
SOCTT, we don’t measure joy, we measure joy AS consumption.
REREAD THAT
“I used to really enjoy reading magazines. I recall occasionally leaving a barbershop or doctor’s office and secretly wanting to finish a National Geographic article I had started on while waiting. I only held back because I didn’t want to look “weird” to the receptionist. Information was really enjoyable because it was so scarce.”
Scott, not to poo-poo your life…
But dude, what a horrible shitty life – I HATED THAT EXACT THING! I used to go sit in the LIBERAL NEWSPAPER EDITORS office, and BE NICE TO HIM, just to read a god damn AP wire, bc I had read every newspaper and magazine available at the community college library.
I got huge amounts of negative suffering bc I did not have EVERYTHING DIGITALLY ALL THE TIME – I want information pumped into my brain while i sleep, I want to UPLOAD Kung Fu into my brain.
Look, it’s these moments where you show the problem facing Economics today:
1. You attribute happiness joy to NOT CONSUMING.
2. You see not traveling to Costa Rica to get MOST of the benefit of Costa Rica AS A BAD THING.
We have to get past this!
You want theories to be falsifiable, ok!
This is what i have been asking for exactly – ASSUME YOUR LOVE OF SCARCITY AND ATOMIC SUFFERING is all bad.
1. Being forced to travel ANYWHERE is BAD.
2. Being forced to WAIT to read something, to consume something is BAD.
3. Dreaming without a direct connection to Internet is BAD.
Now then, we are falsifying your assumptions:
Our task is to find the correct NDPLT that ENDS THE BAD.
That is:
We want to go from here:
Natural reality ————–> computer program Scott
to here:
computer program Scott
We want ATOMIC companies to die, we want to shift consumption of Costa Rica from a horrible terrible human experience full of bugs, and sweaty humans stuck in a stifling airplane full of crying babies to blue beaches and awesome cocktails naked women – and we want people to be there ANYTIME THEY WANT.
So what level of NGDPLT gets rid of travel the fastest Scott?
What gets rid of Scott WAITING?
I know you can do this!
17. June 2015 at 04:17
future note: back arrows are not allowed by WP
17. June 2015 at 05:01
I’ll know the digital revolution is real when it puts a dent in the costs of non tradable goods.
17. June 2015 at 05:29
I guess it depends on one’s perspective of where we are in the process. I mean, there have been tremendous changes brought about by the digital revolution. On the other hand, there still is a real Costs Rica, and “when nature calls” I still have to “answer” physically, not digitally! Now when we get to Kurzweil’s Singularity maybe I can check that one off the list.
17. June 2015 at 05:49
Becky, I agree with you – non-tradeable goods (local service sector consumption will be the last to go) – before that 3D printing (pay for the atoms, not the IP) will drive down the cost of physical assets to $1 per lb.
Which means Uber for Welfare is a great way to maximize human consumption AND it benefits us to get human labor in a welfare state priced correctly – meaning we price labor AFTER assuming everyone must work and everyone’s needs are met.
Gofx, that’s the problem Scott’s not dealing with…
Since you can grok the Singularity (digital consciousness), just as Scott can COMPLAIN that all these pictures (soon VR) of Costa Rica make GOING THERE less valuable…
We are now clearly into my frame:
Today (atomic) headed towards tomorrow (digital).
Scott doesn’t ADMIT that if he is LOSING value in his trip to Costa Rica…
9B other people are GAINING value bc they get most of Costa Rica for free!
Take ANY complaint Scott has about what is “lost” by him in digital space….
Make it a positive number and multiply by 9B.
17. June 2015 at 06:20
Well…the key takeaway from this unusual post is that measurements of inflation are subjective and somewhat arbitrary. Therefore a central bank obsessed with an IT makes no sense.
I keep coming back to it: the Australian Reserve Bank targets an IT band of 2% to 3 % thereby somewhat conceding that there is no true measurable rate of inflation, but between 2% and 3% is good enough. Maybe 2.5% to 3.5% would be better.
Of course, the whole argument gets even sillier when you consider that some people are obsessed with a nominal index of inflation instead of real economic growth.
Maybe the Fed should target 3% unemployment and forget about inflation.
17. June 2015 at 06:20
The reduction on the labor participation rate maybe what society is “paying” for the “free wealth” of the digital revolution. Families have less people working, on average…
17. June 2015 at 06:26
I broadly agree with Morgan’s critique. Basically, you say: “with the digital world “more” begins to seem rather monotonous”, but that’s just an indication that you’ve reached a point of diminishing returns, and that’s got to represent a significant amount of growth.
‘Infovores’ of course gain the most as Morgan says, but even if you’re not one, quality improvements are still on the table. Maybe you only read a few articles a month, but the ones you can find on the Internet are better than the old NatGeo magazine.
17. June 2015 at 06:45
I want to agree with Morgan. I really do. But I am more pessimistic than even Sumner. I am more in the Krugman/Cowen/Thiel school of thought on innovation. What we have is overrated. Give me the hyper loop or actual driverless cars. Give me a space elevator or a cold fusion reactor. Then we can actually get more for less, the definition of productivity.
For now, I am sorry, but all we have with the internet is at best an amazing information sorter and at worst just a cool toy we are all addicted to at the moment.
17. June 2015 at 07:23
1) I am with Michael Mandel that the productivity boom in the 2001- 2004 was the great outsourcing gains. (So all the productivity gains were made when Chinese built stuff went through Long Beach ports.)
2) Probably the biggest sign of “Digital Deflation” is in IT software investment. We are spending less than 10 – 20 years ago but I believe we are getting a lot more. Back in the Clinton days software office rollouts tooks days to train and rarely work as they should have. Today, office software rollouts take a couple of Webex trainings and they 80% as they should.
3) Is some of the ‘productivity gains’ simply the improvement of manufactured goods? It amazeds people to learn that the average US car sales has been ~15 million for about 30 – 40 years. (Lots of up and down years but it has not moved a lot.) Why is true? Cars used to last 8 – 10 years and now they last 12 – 15 years. (This is true for other household appliances too) This does not end up inflation numbers.
In terms of inflation/deflation I am surprised that few economist are talking the huge slowdown in family formation. It seems impossible to increase AD or inflation when most of the richest nations have people getting married and having children until they are 32. The reality is people spending peaks 40 – 45 years old and that portion of the population is shrinking in Japan and Europe today.
17. June 2015 at 07:28
One question for Scott…
Why do you point to NGDP as an indicator of slow growth. I thought your view was that NGDP was something that is totally under control of monetary policy. Shouldn’t the slow NGDP growth be telling us that even now, monetary policy is STILL too tight?
17. June 2015 at 07:32
It Morgan Warstler day on Internet! Here’s Bruenig DESPERATELY bailing to try and ward off better than yesterday analysis:
http://www.demos.org/blog/6/16/15/people-arent-better-income-trends-show
Liberal Roman,
I say to you what I say to Bruenig (and Scott). The Internet GAVE us gay marriage. Just like refrigeration gave us women’s vote.
Mark Twain said “Travel is fatal to prejudice, bigotry, and narrow-mindedness, and many of our people need it sorely on these accounts. Broad, wholesome, charitable views of men and things cannot be acquired by vegetating in one little corner of the earth all one’s lifetime.”
The Internet alone topples backwaters with a speed and breadth most of us CANNOT believe.
Scott spent his life, like did wondering why drugs weren’t legal and gays couldn’t marry and and and…
I however knew as an infovore, that if we could just FLOOOOOOOOOOOD every backwater with all the info at our fingertips, then everyone would be more like me and Scott!
And it works! Technology MAKES us Libertarian.
Anyway back to Roman,
Ever see those movies where the protagonist wants wants wants something and they go thru long journey and they finally get it and htey never want it again?
There was a book about Christian Science in 1970’s called Jonathan Liviingston seagull – that capture my argument to you:
He wanted to go faster and faster, move physically thru atomic space faster than any seagull, any being ever had before – and then – he realizes atomic space is not to be overcome, bc he’s already there, where he needs to be.
The point is you want a flying car!
i give you flying on backs of dragons firing lasers at goblins to protect the Swedish bikini team.
If you so NOT admit the MASSIVE gains that come from consuming Costa Rica without going to Costa Rica – HOW THE GELL CAN YOU CAL YOURSELF AN ACCURATE MEASURER OF HUMAN CONSUMPTION?!?
Scott MOANS that his trip is less valuable, bc 9B others got it for free.
Does he MOAN that everyone will now support gay marriage, even though they grow up in Saudi Arabia? BUT, BUT, BUT Scott wanted to go to a place where backwaters still exist!!!!
This is mind boggling.
Scott when you find yourself agreeing with Matt Bruenig – I mean c’mon.
17. June 2015 at 07:45
Inflation is much easier to nail down when we limit it to the cost of living; housing costs X and food costs Y. Add up the price of those things, compare them year over year, and you’ve got inflation.
It gets complicated here in the first world where we have more things that are good for more than just living. We can attempt to define it as something like inflation = Year-over-year utility per capita / rgdp per capita but we can’t really measure utility and it’s hard to measure rgdp for things that cost money.
I’d argue that the first definition is the only one that the public cares about; the public worries about the cost of supporting their families and wants a COLA to accurately reflect any increases in that cost. We economists can debate the second definition all we want but let’s at least agree that it’s mostly academic rather than practical.
17. June 2015 at 07:48
@Morgan W – dude we are counting on you to do cyber-smell and virtual reality porn. I’m counting on five fingers…
17. June 2015 at 08:49
@Morgan,
Sorry, but you are starting to sound like a hippy. Next thing you’ll be telling me I need to stop worrying about concrete things like Gross Domestic Product and start worrying about Gross Domestic Happiness.
Reading your response made me realize of the PERFECT comparison for the Internet. The printing press! The printing press certainly changed quite a lot. It’s basically undeniable that Gutenberg’s printing press led to the Reformation. And certainly allowed people access to information that they had never imagined they could have before. And no one is denying that the printing press was great and perhaps a necessary precursor to all the political and economic advances that followed.
BUT, and this is a huge BUT….it was about 300 years between the invention of the printing press and the Industrial Revolution. 300 years between the printing press and actual sustained per capita GDP growth.
So I think we are in that 1400-1700 period now where we have many new diversions, entertainment options and sources of information, but that’s not really giving us actual material gains.
17. June 2015 at 09:47
i sort of wonder how we can try to justify saying that living standards are better today than in the past based on comparing the technology used by each? are we thinking that the peasants in the dark ages are better or worse off than the slaves in the Roman empire because of their technmolog?
17. June 2015 at 10:22
Kevin. Yeah, if I cared about popularity. But I long for anonymity, for having time to do fun things like reading fiction.
Dennis, Things may have improved, but not my email (Outlook). I frequently lose email messages I just spent 20 minutes typing. And my computer (an expensive iMac), freezes up about once a week, and I have to turn off the machine and start over.
Christian, OK, but is there data to back that up? Is life expectancy rising more rapidly in recent years? Are people suffering from less pain?
Daniel, Yes, it’s NGDP growth that matters.
Chris, I’m not sure younger people realize that if this stuff had never been invented, they’d get much more enjoyment out of simpler forms of entertainment, which now seem boring (because the more advanced stuff does exist.
17. June 2015 at 10:29
Guys,
Scott has already admitted he’s wrong. Now it’s just a matter of measurement.
“Travelling used to be about discovery, strange new worlds you’d never seen. Now it’s “Yup, Costa Rica looks just like it did on the internet while I was planning the trip.””
How much worse is Scott’s vacation to CR?
Let’s say he spends $3K on the vacation, and he’s already consumed 20% of that looking at it online before he makes the jump.
Ok, so now we know the value ($600) of EVERYONE ELSE who doesn’t go to CR and just consumes in on the Internet.
Let’s say 10M others consume that $600, ok! now we have our measure – we have an extra $6B in REAL CONSUMPTION, that is not being counted correctly.
Roman, I’m not a hippie!
But the consumption shifts are like this:
Printing Press (600 years) – which made Scott WANT to go to CR
Internet (20 years) – which TOOK AWAY 20% of Scott’s value from trip to CR
See how we had to spread the gains of printing press over 600 years, but we’ve crammed in far more into Internet’s 20? See how the gains of Printing Press are REVERSED by Internet – by making us NOT WANT TO TRAVEL.
Frankly, there is NO internet invention that doesnt scratch a very basic human itch: to not go anywhere.
Amazon, Ebay, Napster, Google – it doesn’t matter – all of it, promises the user LESS UNNECESSARY ACTIVITY. Sit on your ass.
Now look at VR, where the 20% slash to Scott’s joy, moves us toward 50%, 60%, 70% until finally Scott’s in his pod, and never needs to go anywhere.
But let’s get this straight: Scott already ADMITTED sizable value to taking an Internet trip to Costa Rica.
We don’t have to discuss if I’m right!
We just have to measure how much Scott loses in his example.
17. June 2015 at 10:41
Ray, I was talking about dark matter before it was cool.
Morgan, You’ll live to see the Brave New World of VR, I won’t.
Ben, Yes, don’t get me started on central banks targeting inflation.
Collin, I totally buy the cars are cheaper argument, google my post on honda and acura
Liberal, It’s complicated. The Fed targets inflation at 2%. So when RGDP growth slows, NGDP slows too. But yes, you can argue money is still too tight, at 3% NGDP growth. However we are much closer to equilibrium than a few years ago, our problems are increasingly supply-side.
Morgan, Great quote from Twain, but you missed the point on Costa Rica. I don’t want them backwards, I just want to be surprised by the look of the place. I want different, not backwards. I’d love to go to Japan because it’s different, not because it’s backwards.
Last summer I had a family vacation in Azores and Madeira. It was great because I had no idea what those places were like.
Randomize, I frequently argue that the public has no idea what economists mean by inflation—that’s also what you are saying–I agree.
Liberal, Of course Morgan is a hippie, he lives in Austin. Read Republican Party Reptile, that’s Morgan.
17. June 2015 at 10:44
Morgan, Consider yourself lucky. I was this close to doing this post at Econlog. (Holds thumb and forefinger close together). 🙂
17. June 2015 at 12:11
That Martin Feldstein article does do a good job of bringing home the absurdity of trying to measure an economy-wide rate of inflation. But then he doesn’t relate it at all to his policy recommendations, which are the usual supply-side ideas. (Not that those ideas might not boost real growth, just that whether it’s a good idea to do them doesn’t have a thing to do with the measured rate of inflation.)
It’s all a little beside the point. Morgan (and Feldstein) may be right that the “real” growth rate of the economy is underestimated by the official inflation statistics, but that’s not relevant at all to macroeconomic stabilization. For that, the only thing that really matters is nominal wage growth.
17. June 2015 at 12:12
I remember hearing most of the increase in RGDP 1995-2004 was from retail, not from tech. It makes sense. Even now, a very low percentage of income is spent directly on computers for personal use. So the increased efficiencies on the back-end, whether for cheaper lumber or cheaper food, matter more.
I feel extremely dumb as far as this basic economics question, but isn’t RGDP also tainted somewhat by the capital vs. labor question? By that, I mean any form of labor can choose whether to invest part of their income or to spend all of it.
Krugman and many others have pointed out a greater share of RGDP has gone to capital, but what if that merely means the preference has gone more from consumption to capital? The economy produces more because you have more capital working alongside labor, but then GDP/capita does not really accord to living standards of a laborer per hour of work. More of that GDP is going to capital.
The higher capital income/GDP ratio also does not necessarily indicate higher corporate profits through monopoly/oligopoly/monopsony/uh…olipsony? bargaining power. Even if real wages diverge from real GDP/capita, the divergence does not indicate somewhere capital is robbing workers. Furthermore, real wages aren’t a great tool because it’s very tough to find if real wages indicate total compensation per hour, as more wages go to health care and other fringe benefits.
These seem like very naive and dumb questions to ask, so I apologize for that.
17. June 2015 at 12:29
Scott, you write:
“While I’m impressed by an explanation that’s as flexible as a circus contortionist, I’d prefer something that isn’t consistent with any possible state of the universe. I’m no Popperian, but I like my theories to be at least a little bit falsifiable.”
Can you provide a concrete example or two of future data which would falsify your theories (at least a “little bit”)?
Also, would you describe any current macro economists as Popperian? It seems to me like stating ahead of time (for the whole world to check) precisely what sorts of future data sets would falsify your hypotheses can only be a good thing. Isn’t there a tendency for people to fall in love with their own hypotheses otherwise?
17. June 2015 at 12:50
Scott (continuing my above):
1. Are you not a Popperian because of the nature of macro data itself? Could improved data (if this is possible) make being a Popperian feasible? Or do yo have a philosophical problem with being Popperian regardless of other factors (factors such as the quality of macro data)?
2. What do you think other macro economists think about this? Do you imagine they largely agree with you or not?
17. June 2015 at 12:55
Scott,
Off-topic, but the following quote is too good not to share:
“Think of the U.S. economy over the next decade as an automobile driving from Boston to New York. Let the average speed of that automobile over the route represent the average level of employment over that decade. And let the dispute over the direct employment effects of X be represented as an argument over whether there will be a head wind or a tail wind as the car makes its way along the interstate. Then assessing X’s overall job impact is like predicting how the extra wind will affect the car’s speed. Job-counting exercises do this by assuming that nothing else changes-in effect, they assume that the engine in our car will receive exactly the same flow of gas that it would have been given in the absence of any wind.
Nobody would think that this was a sensible procedure for predicting automobile speed. After all, cars have drivers, and drivers are not passive-they adjust the flow of gas to achieve a desired speed. I tend to drive interstates at about 63 miles per hour-above the speed limit because I’m always in a hurry, not too far above because I would prefer that the police chase someone else. A five-mile-per-hour head wind or tail wind will not change that average speed; I will simply offset the wind by changing the pressure on my gas pedal.
The U.S. economy also has a driver: the Federal Reserve. Every six weeks or so the Federal Reserve’s Open Market Committee meets to decide on a target range for U.S. interest rates. That choice has a far more powerful impact on the unemployment rate than any trade policy. Moreover, it is a choice that responds to economic conditions; the decision to raise or lower interest rates represents a trade-off between the Fed’s desire to raise employment (drive somewhere) and its fear of inflation (a speeding ticket). The Fed often miscalculates and ends up with more inflation or less employment than it wanted, but right or wrong the Fed’s actions are the most powerful determinants of job growth in America.”
Can you guess the author?
17. June 2015 at 13:25
O/T, but don’t you have to love statements like this?
The Fed is forecasting that the Fed will fail to hit the Fed’s target.
17. June 2015 at 13:30
I need to stop worrying about concrete things like Gross Domestic Product and start worrying about Gross Domestic Happiness.
The world desperately needs a measure of Gross Domestic Utility, alas it can only be found in the eye of the beholder.
17. June 2015 at 13:31
“With the atomic world, there was a sort of logic connecting more stuff with higher living standards. More food, a washing machine, a TV and telephone. But with the digital world “more” begins to seem rather monotonous. I used to really enjoy reading magazines. I recall occasionally leaving a barbershop or doctor’s office and secretly wanting to finish a National Geographic article I had started on while waiting.”
“I only held back because I didn’t want to look “weird” to the receptionist. Information was really enjoyable because it was so scarce. Now there’s a sort of infinite magazine at my fingertips. And it’s all a bit too much. Travelling used to be about discovery, strange new worlds you’d never seen. Now it’s “Yup, Costa Rica looks just like it did on the internet while I was planning the trip.” And what about the negatives, the constant annoyance of your computer freezing up, or losing a long email that you had almost finished typing, and hadn’t saved.”
I can totally relate to that. As a teenager in the 80s I had the same experience when my Mother would take me to the dentist. It was worse in a way because I had the feeling that other kids thought reading was uncool. So I was even more embarrassed to be seen hanging out in the office reading.
But I really feel like this is in a way generational because I feel totally gratified to have an ‘infinite magazine at my fingertips.’
More than that, an infinite library at my fingertips. With Amazon Kindle any book that comes to mind can be downloaded and read on the spot.
Obviously you and I have different preferences on this because to me the pre IT way was veryfrustrating. There was all this information out there and it was so difficult to get at. I much prefer being able to realize my wants in this way.
The reason I say it’s generational is that all the guys of a certain age-no offense intended-like yourself, Tyler Cowen, and yes, even Krugman seem to be somewhat dismissive of the IT revolution.
While I find myself-rather surprisingly-agreeing with Morgan here-and he and I are basically from the same generation.
Now I have no idea how or if you should even try to measure this in economic terms. But it seems to me having an infinite library at the fingertips is worth something pretty significant.
It has totally revolutionized our society; apparently 25% of marriages today begun on the Internet.
You yourself have met many econonerds who share your interests that you never would have achieved otherwise. In fact this whole Market Monetarist thing could never have gotten off the ground being argued solely in the academic econ journals-the AER, etc. So I think maybe you’re underestimating it.
17. June 2015 at 13:37
P.S. My father is like the last person I’m aware of who doesn’t own a cell phone period. He’ll admit that there could be some benefit-at least in emergencies. But always finds some excuse to put buying one off.
17. June 2015 at 15:35
Seems to me somebody’s got to mention the creepiness of (voluntarily?) allowing the aggregators 24 hour access to every detail of your life.
FB and Google don’t just follow you wherever you go, they functionally live inside your home, root through every drawer and closet, every bit of your trash, and even every corner of your mind, constantly analyzing all of it for the express purpose of “monetizing” you.
That’s the heart of the Digital Economy, and the question I wonder is, compared to the past, is it a departure in kind, or merely extent?
17. June 2015 at 15:37
For me digital is absolutely crucial part of my life and I almost cannot imagine ever going back. If I could choose between a week in 5 star hotel or in some random motel with broadband I choose the latter.
I read articles and blogs. I play computer games and watch youtube videos. I listen to music via Spotify and watch TV shows over internet. I know in general what my friends are doing thanks to facebook. And I have IT related job, so IT is literally feeding me.
Plus internet has a tendendcy to improve whatever you were doing before because you have all the knowledge at your fingertips. You get to know about everything. You can find inspiration for what to cook, you can check user reviews of schools for your kids, you can hook up and exchange your flat with somebody in France for a week. You can buy used things on ebay and save a lot. And I barely scratched the surface.
There is a reason why people in developing world prefer cell phones and internet over indoor plumbing. If I was in their shoes I would have the same priorities.
17. June 2015 at 17:03
I find this proposition shallow and not at all appealing.
1) Scalable production doesn’t equal greater output – 1 article published on the web and consumed world-wide by 1 billion people is still just 1 article of real output.
2) There is no such thing as ‘free’ – all the ‘free’ content we consume is offset by increased spending elsewhere. This is self-evident when considering that ‘free’ content is monetized via advertisements. If consumption elsewhere were not increased, the entire concept of digital advertising and ‘free’ content is inherently unsustainable (and this entire conversation is irrelevant).
3) Utility is not production – yes of course the internet increases quality of life, so what? So does fresh air and greenspace. It would be absurd to argue that national parks lead to deflation or the understatement real output.
17. June 2015 at 17:05
dbeach, Again, there cannot be a truth of the matter regarding statistic X, when no one has adequately defined statistic X.
Matt, Part of the extra return to “capital” is income going to owner-occupied homes.
Tom, Philosophers that know much more than I do have a lot of problems with Popper, and I find their criticism to be quite persuasive. I’m a pragmatist, which means I judge theories using many different criteria, of which one is falsifiability.
My sense is that most economists know very little about philosophy, and their views on methodology aren’t worth paying much attention to (with a few exceptions like Deirdre McCloskey.)
SG, Don’t know, but it sounds like something Krugman would write.
Mike, But it’s not really kosher for me to judge the role of IT in 312 million Americans lives by looking at my own somewhat atypical trajectory.
AL, Yes, and the NSA also wants access to all that stuff.
JV, In airports and on airplanes I see all these people with laptops, and then there’s me, just reading an old fashioned book.
17. June 2015 at 17:45
ssumner: “Ray, I was talking about dark matter before it was cool.” – show your work. I don’t believe you. Dark Matter and the Intl Payments solution by economists Ricardo Hausmann and Federico Sturzenegger was published around 2005.
17. June 2015 at 18:20
These 2 comments pretty much sum up my sentiment.
J.V. Dubois:
“For me digital is absolutely crucial part of my life and I almost cannot imagine ever going back. If I could choose between a week in 5 star hotel or in some random motel with broadband I choose the latter.”
“I read articles and blogs. I play computer games and watch youtube videos. I listen to music via Spotify and watch TV shows over internet. I know in general what my friends are doing thanks to facebook. And I have IT related job, so IT is literally feeding me”
And Morgan:
“I got huge amounts of negative suffering bc I did not have EVERYTHING DIGITALLY ALL THE TIME – I want information pumped into my brain while i sleep, I want to UPLOAD Kung Fu into my brain.”
Amen.
17. June 2015 at 19:59
Scott, thanks. You write:
“I judge theories using many different criteria, of which one is falsifiability.”
Isn’t falsifiability simply a prerequisite to even be able to judge a theory? (e.g. it could be both falsifiable and false)
17. June 2015 at 21:18
[…] makes many good points, but this is my […]
17. June 2015 at 23:19
Dustin, I don’t agree with any of your points.
1) An article read by a billion people is certainly more output than the same article read by a thousand, in the sense that matters, that is, are people receiving utility from it? I mean what’s the standard? If I sold a billion access codes to read the article, surely that’s output, yes? Then how is it not output if people read the article for free?
2) Demonstrably false. Wikipedia is free, it has no ads at all. This very site has no ads at all. Some people just produce web content because they want to. It’s only unsustainable if you think people will cease wanting to communicate with one another.
3) This is where we get into the folly of measured GDP == production. Suppose I paint a masterpiece, the art critics rave over my work. I could sell it for $5 million. If I did, it would show up in the GDP figures. But I do not, instead giving it as a gift to my one true love. Do you really believe that I produced no output?
Again, I still think this argument is irrelevant to the subject of this blog, economic stabilization. It does not matter for the cause of smoothing output growth what the “real” *level* of output is or how much utility consumers are receiving from that output, and it especially does not matter how much utility they are receiving from things — physical or intangible — that they get for free. It only matters if the nominal growth of actual monetary expenditures maintains a steady pace.
18. June 2015 at 01:20
dbeach,
“Again, I still think this argument is irrelevant to the subject of this blog, economic stabilization. It does not matter for the cause of smoothing output growth what the “real” *level* of output is or how much utility consumers are receiving from that output, and it especially does not matter how much utility they are receiving from things “” physical or intangible “” that they get for free. It only matters if the nominal growth of actual monetary expenditures maintains a steady pace.”
Steady pace? yes!
BUT!
The question I’m asking is slightly different…
If we have 4% NGDPLT instead of 5%, and then overtime move to 3% NGDPLT and then in 20 years 2%…
I’m suggesting this may INCREASE THE SPEED of the shift from digital to atomic.
And anything that doesn’t INCREASE THE SPEED of the shift from atomic to digital is:
1) immoral
2) anti-utilitarian
My fear is that since Scott doesn’t look at 2008 till today as BEST TIME IN HUMAN EXISTENCE for real consumption, he might be advocating policies that kept atomic economy open longer.
—–
Think about Atomic / Digital Economy like Old Greece / New Greece (Greece that lives and operates like South Carolina)…
SINCE New Greece is BETTER, anyone advocating ANY policy that allowed Old Greece to exist LONGER – this isn’t the best economic policy for Greece.
And Scott should be able to grok the falsifiability I’m asking for from him!
Just assume last 7 years have been best years for real consumption in American history.
Now, given that, why would having higher NGDP have bankrupted more retail, closed more office buildings, and created more Uber like businesses?
It’s a good true check of Scott’s underlying assumption.
18. June 2015 at 02:58
Morgan Warstler wrote:
“If we have 4% NGDPLT instead of 5%, and then overtime move to 3% NGDPLT and then in 20 years 2%…
I’m suggesting this may INCREASE THE SPEED of the shift from digital to atomic.”
I don’t think so. The changes that a tigher monetary policy might drive are not necessarily the kind of changes that you would want to see. (See the Great Depression and FDR).
18. June 2015 at 05:05
One weird aspect of the digital revolution, is how quickly the market settles on a limited number of companies. The markets settle on a monopoly very quickly. In terms of Facebook, they did nothing completely new, are very profitable, markets don’t have high entrant costs, captial costs are not high, and, yet Facebook almost has a monopoly over the market. You could say the same thing on Google, Twitter, Apple, Amazon and Microsoft.
Maybe digital companies simply have an insanely low marginal cost curve but the cost to users are significant learning a digital company are high.
18. June 2015 at 05:05
Scott I usually don’t feed the troll on this site, but since you blogged it let me also point out: there was a large increase in communication and entertainment through the advent of television but nobody thought it was missing in the GDP numbers. What is the difference except scale? Same with radio, telephone, talking pictures…just entertainment. And sadly these techno trolls can never put forth a more coherent argument other than declaring loudly that poor people are really happy in their poverty due to cell phones. And lying about what people think, which is why he is nothing more than a troll.
18. June 2015 at 05:11
dbeach,
1) Consumption does not equal production. If we freely view an artist’s painting, does each viewing count toward GDP?? If I drive along a road a thousand times, does each pass count toward GDP?? Or if 10 friends play my guitar, does each session count toward GDP?? This would all be incredibly bizarre and so greatly water down the concept of GDP to the point of it not being worth a discussion.
2) Wikipedia is a not for profit, and yes I’m with you that creates challenges in fully measuring real output, as does housework. However, this is hardly a new phenomenon, nor is it unique to or the standard for the technology era. Most ‘free’ stuff is supported by plain old ad revenue – and the point still stands, the $ are spent elsewhere. To drive this back to Scott’s post – would folks be OK with stagnant or reduced nominal income? Nope not at all. We want a growing nominal income to spend elsewhere … a nicer car, a nicer house, organic food, a carbon road bike, more trips to Disney Land, a nicer wristwatch, etc… Technology does not – in any way – reduce AD.
3) Are we really disagreeing? Utility does not equal production – real or nominal.
18. June 2015 at 05:41
@Scott – first try, I’m impressed!
18. June 2015 at 06:08
@Benny Lava – “sadly these techno trolls can never put forth a more coherent argument other than declaring loudly that poor people are really happy in their poverty due to cell phones” – are you aware of the studies–and my own first hand experience living here in the Third World Philippines–that poor people indeed rather have a cell phone than indoor plumbing, all things being equal? It’s a fact.
18. June 2015 at 06:32
“What is the difference except scale?”
Benny Lava – not dunb, but so close minded he can appear dumb.
The difference is DIGITAL. Of bits and bytes. Add in Moore’s Law and network effects (each new user in network increases value of network exponentially). Each new copy has zero cost. Each new user wildly increases utility.
What the Internet did to Entertainment…. Benny you have not yet seen 3D printing do to manufacturing… but you WILL see everything physical cost you $1 per pound. Benny, you WILL see classrooms and campuses be UNABLE to compete with video / data instruction. You WILL see VR capture 50% of the joy of a actually going someplace.
I know this as clearly as I knew Internet would topple Entertainment. I personally did the toppling, Benny. So when I look at thing and measure it out, I’m not just hand waving, I can tell you exactly what forces in tech market already have int he roadmap – stuff the R&D guys have already patented.
Today I work feverishly on a stealth newco that replaces all of state and local govt with a single app. What Xerox does for $20M a yr, we can do for far, far less, and it takes about one hour to build the complete and functional mobile solution. I mean it Benny, one hour.
But let’s get back to 3D printing Benny…
I just spent almost $300 for 3 large plastic pieces that fit underneath the front end of my wife’s car (the dealer charges $500), the Internet has already saved me $200. The waste of human labor that goes into making these things storing them, shipping them – when for $100 (and huge profit) I will be able to head over to neighbor’s local 3D print shop and pick up the parts I ordered, and at scale the price should fall. At scale, a dude in truck with 3D printing onboard pulls up outside and not only prints but installs whatever plastic (then rubber and metal) parts people need.
It’s hard to understand your point Benny, the Internet is not “Entertainment” the Internet DOES TO EVERYTHING what it has done to Entertainment – makes everything that used to cost money FREE for most of the planet. 80% of the people on earth pay ZERO for music and movies Benny, except the ones who can afford the atomic experience Benny – thats means theaters and concert halls, and subscription based content for the top 20%. And Benny the bottom 80% they get to see EVERYTHING you get to see!
Personally I advocate (and did so when officially advising RIAA and SAG) just making everything that can be copied free – bc you cannot “own” the digital Benny. Ownership requires atoms Benny. Here you’ll like this… Digital Socialism, Atomic Capitalism:
http://www.morganwarstler.com/post/35224055375/digital-socialism-atomic-capitalism
18. June 2015 at 06:40
OT- June 5, 2015 post by this provocative physicist who dabbles in economics. Already he’s upset the status quo in economics, who claim they don’t understand his blog.
Market Monetarism (MM) is not falsifiable: see http://informationtransfereconomics.blogspot.com/2015/06/falsifiabilite-simplicite-succes-ou-la.html
The other aspect of it is that market monetarism suffers from “no true Scotsman” disease with its predictions. If the Fed doesn’t achieve its inflation target, it must not have wanted to achieve it. Market monetarism can be proven wrong (a better word is outperformed) by a more concrete, falsifiable model that gives empirically accurate predictions. Basically if there is a theory that works better than handwaving about central bank mindsets and targets, it wins.
(And, btw, I, Ray, posit a concrete, falsifiable model is that the Fed follows the economy. Show me when the Fed jump-started the economy with an exogenous interest rate change, and I’ll concede defeat. But it’s not happened, statistically, and as I said in another post only with a mere 60% confidence interval (rather than the conventional 95%) can you show the Fed has any short-term influence over the market. )
18. June 2015 at 06:44
Morgan makes a good point — I’ve saved at least a thousand dollars this year by Googling how to fix things. GDP fell, but utility was increased.
What is Wikipedia worth? Some people have never heard of it, others find it incredibly fascinating. Eye of the beholder.
Anyways, it’s not a new problem in economics, as long as GDP has been around it’s been known that it does not capture utility, nor is it intended to — for instance, GDP rose during WW II even as we rationed, conscripted, and suffered/inflicted mass murder. Living standards are not the same as GDP.
3D printing will always be a niche application, though, because of the materials limitations — it would be nice to live in The Diamond Age but the cheap matter compiler isn’t going to happen, there’s just no way to make that energy balance work.
18. June 2015 at 07:28
TallDave
NGDP is not affected by the savings unless you put the $ under your mattress – otherwise you spent the $1K elsewhere or saved/invested it and someone else spent it
Reduced prices for goods and services does not impact AD
18. June 2015 at 08:14
TallDave
The cheap matter compiler 🙂 I’m not expecting that.
I am expecting any object made from one material to be 3D printed and I’m expecting most factory manufacturing to move towards publishing 3D printable parts (open source), or being overwhelmed by 3D scans (pirated copies). The idea being that buying a thing automatically comes with a license to as many copes of those parts for yourself that you want – at a minimum. IKEA and Ford and everyone will go along with this….
Speaking of furniture, I think 3D printed wood working (robot router) will cut the legs (terrible pun) out from under much furniture manufacturing in another 10 years.
In general, the DIY movement in atomic space, I think it’s literally at the 1995 web page…. as an example: I’m looking at $2K used laser cutter for my wife’s leather bag making company – I think over the next 15 years, the quality of work you find in DIY anything, will get very near what you see coming off factory floor. Digital printing of fabric is just as exciting.
Electronics – there’s simple no need for 3D printing – but I think we’ll see far more programmable stuff.
18. June 2015 at 08:50
Scott, VR!
http://gizmodo.com/oculus-touch-hands-on-so-damn-good-1712031397
18. June 2015 at 10:44
This is one of great things about NGDPLT. It is easier to measure than inflation or GDP.
Just a note people used to line up on payday to buy music now pandora.
18. June 2015 at 14:08
It’s highly annoying to hear the pro-tech-deflation side misrepresenting the nature of the side of the skeptics. No one is seriously suggesting that the internet and tech generally have not made very big quantitative and qualitative improvements to quality of life. The argument is over the relative rate of improvement 2005-15 vs 1995-2005.
If you are in the camp of believing that the rate of productivity growth in the economy is being seriously understated because of internet/tech-related qualitative improvements, then you still need to answer the question as to why it showed up in productivity stats 1995-2005 but not 2005-2015? Do you posit that there is some qualitative differnence in the two periods that makes one kind of productivity increase measurable but not the other kind? Is there some constant unobserved increase that when added to measured productivity makes the relative change smaller (i.e. did the real productivity growth go from 4% to 3% instead of 2.5% to 1.5%)?
If you are going to do this you need to have a clear idea what the state of the internet/tech was in 2005. In 2005, residential broadband had become ubiquitous, Google had already crushed Yahoo in search dominance, Netflix was already a profitable public company mailing 1 million DVDs daily, iPods were selling at about half their eventual peak rate, iTunes was on version 6, weather/traffic navigation updates were a premium product you could get on your TomTom, “Blogosphere” had been a word for 3-4 years, Wikipedia was 4 years old and doubled its article count from 450k to 900k, every major financial institution had online management tools, World of Warcraft was already more profitable for Viacom than the entire Sony movie division and it was common for AAA game titles to outgross action movies domestically, DVRs were a standard option for cable, Ebay had been in the S&P 500 index for 3 years, Amazon’s book sales had long since surpassed Barnes & Noble’s, “The Long Tail” had already been published as a magazine article and was being expanded into a book, and the National Retail Federation coined the term “cyber monday” to describe what had obviously been going on for several years. In short, a lot of what we think of as the core functions of the internet and changed entertainment consumption habits were already in place and entering maturity.
The big internet/tech differences 2005-15 revolve around 3 trends: the rise of social networks/media, the (related) rise ofthe ability of amateurs to produce their own entertainment content, and the shift from fixed (PC & TV) to mobile devices as the vehicle for the consumption of media. In 2005, Facebook was limited to about 3 dozen elite colleges, Twitter was a year away from being founded, Youtube had just been founded and would not be bought by Google for another year, the iPhone was still in development, the Kindle was 2 years in the future, and the iPad was 5. Obviously these trends represent real improvements, both qualitative and quantitative, but do they really measure up to what was done 1995-2005? I would say no.
Compare at the anecdotal level. Today in 2015, you can stream pretty much any TV show ever aired at any time you want in any location with wireless reception. Not just produced TV, but anything on youtube as well. It’s fantastic. In 2005, you could watch almost any semi-popular produced show you wanted, but you had to either buy the DVDs, wait 2-3 days to get them in the mail from Netflix or tell your DVR to record it when it aired, and you could only watch it on your TV. Big improvement? Cheaper? yes of course to both. But what was the relevant comparison point in 1995? That was the year DVDs were first invented. There were maybe a dozen movies you could get on it, and zero TV shows. If you wanted to watch a TV show that wasn’t being aired, you were 100% SOL. If you wanted to time-shift a TV show, you had to either just know when it was on or get and search a (printed) TV guide, and then program your VCR to record it onto a cassete tape. Your on-demand movie selection was limited to what was in stock at your nearest Blockbuster or what tapes you’d purchased. Let’s be real- what’s the bigger improvement, 1995-2005 or 2005-2015? It’s not even close.
Go and repeat this exercise for most of what you do on the internet. With the exception of social media, the answer is usually “in 2005 it was definitely there, but in retrospect was a bit clunky, a bit limited, and you couldn’t do it on a phone. In 1995 it was utterly unheard of”
18. June 2015 at 15:30
“If you are in the camp of believing that the rate of productivity growth in the economy is being seriously understated because of internet/tech-related qualitative improvements, then you still need to answer the question as to why it showed up in productivity stats 1995-2005 but not 2005-2015?”
Two answers:
CLOUD: If you had any idea HOW MUCH I paid Compaq / Digital in 1999 you’d NEVER mention thus again.
MOBILE: Computers have to be replaced every 2 years, if not 1 year, this was unimaginable in 1999.
So pls, stop with your shallow analysis, just accept it. It makes you look naive.
18. June 2015 at 15:32
Ray,
You realize you demonstrated my point, right?
18. June 2015 at 16:21
@Benny Lava – no, I did not realized it. Please explain in small words. No offense Benny but you write like an economist.
Me: Scott, you say X, but facts say X’ (i.e., the opposite).
Scott: You demonstrate my point.
Me: Huh?
And see my last post upstream why MM is not falsifiable (hence false, like mythology). Sumner indirectly seems to try and answer this accusation in the next blog entry where he discusses which model, MM or Keynesianism, describes better the effects of austerity at the ZLB. Recall a model that explains the facts more simply than another will ‘win’, using Ockham’s Razor, over the other model. MM = metaphysics.
18. June 2015 at 16:55
Ray, I don’t carry a tape recorder around with me and tape all my conversation. Please don’t hurt my feelings by saying you don’t believe me.
Tom, I don’t think so. If I have a theory that WWII was caused by “nationalism” or caused by “The Great Depression” I really don’t see how those theories could be falsified. I’d judge them by other measures, like coherence, and consistency, and plausibility of the components of the argument.
Collin, I always assumed it was “network effects” People use one company because others are using that company.
Benny, You are welcome to feed trolls in this post.
SG, I know how he thinks, and that he’s a master of metaphors.
Ray, You said:
“Show me when the Fed jump-started the economy with an exogenous interest rate change”
If only you knew how funny that was.
And I have entire posts on how MM is falsifiable. Have you forgotten Krugman’s 2013 “test of MM?
Matt, Good comment.
18. June 2015 at 21:49
Morgan-
I’m going to be generous and assume you didn’t quite get what I was trying to say. It mainly revolves around the macro statistical issue of counted vs uncounted productivity. I’ll address both things you mention
Cloud- Cloud computing is essentially the perfect example of a technology improvement that would get counted by existing productivity measures. Why? Cloud computing is mainly an intermediate good. The overwhelmingly majority of cloud computing services are used by for-profit entities in the business of producing something else. Productivity gains that occur inside of a for-profit entity are by far the easiest ones to count because there are cashflows associated with them. Typical example: Large Widget Company makes (duh) widgets, and like all large companies has big IT expenses, both for mundane stuff like email systems, customer-facing websites, data & document management, etc and also for stuff like the computational needs of the R&D dept tasked with improving widget quality. One Day the CFO and CTO of LWC realize in a meeting that a large chunk of their IT expenditure is spent on computing power which in point of fact goes mostly unused, and decide that outsourcing some of the computational/storage/webserving systems to IBM or Amazon would save them approximately 1 metric asston of money. LWC can now make the exact same widgets at a lower cost, and maybe the availability of cheap computing power allows the R&D dept to make improvements it otherwise wouldn’t have been able to manage. Productivity gain achieved! However, because LWC is a for-profit entity, it has a cashflow statement from which the productivity gain can be imputed, and this imputation will get counted up in macroeconomic statistics. Lower production cost of widgets can be realized in several ways, all of which are mutually compatible:
1) If lower widget cost turns into lower widget price, this gets measured in the price level
2) If LWC has the pricing power to keep its widget price constant, then all else equal profit will increase. But, increased profit will get captured and counted in one of the following ways:
2a) Labor will capture some share as higher income, which will show up in both personal income statistics and expenditures on whatever it is workers buy with their $
2b) The owners of the capital structure (equity + debt) will capture some. This won’t show up in wage statistics, but it will show up in income statistics and also again in expenditures on whatever it is the investors buy with their $
2c) If the company spends he profits reinvesting in the company’s business, it will show up in capital expenditure statistics.
2d) If the company just leaves the money in its bank account it will either show up as excess reserves on the Fed’s balance sheet or it will get lent & invested per the normal course of a banking system.
3) The only fuzzy place here is if the R&D group is able to improve the quality of widgets in a way otherwise unavailable. In this case, the gains will theoretically show up in real price level calculations when some gnomes in the labor dept calculate the hedonic adjustments for widgets. The untrustworthiness of such calculations is in part the raison d’etre of this entire blog, but people do actually attempt to estimate it. Arguing that this quality improvement is not measured is straight up false. Arguing that it is mismeasured is entirely legitimate, but unless you know in technical detail how said gnomes calculate said hedonic adjustments or have an alternative measure (numbers, not hand-waving) that demonstrates whatever they are doing has to be wrong, then you’re just blowing smoke.
Mobile: First and foremost, my argument about mobile is that the vast majority of mobile represents doing things you could already do on a laptop/pc, just that you can now do them from anywhere. I concede this is a legitimate real gain that is probably not measured well (any parent who has ever been on a roadtrip will swear on their kids’ college funds as to the life quality improvements of tablets). I don’t think it’s so great as to amount to an uncounted >=1% CAGR for 10 years on macro-level economic statistics. On the counted side, some of the improvements will actually get into statistics. In 2005 you’d have bought a $1200 laptop and a $150 dumbphone and expected them to last 3 years ($450/yr amortized). Today that can be replaced by a $450 tablet and a $300 smartphone which have an expected life of 1.5-2 years ($375-$500 per). We can cuff that as roughly flat given the highly approximate prices I’ve used, which would represent deflation in constant dollars (i.e., relative to the rest of the goods and services in the economy). “Aha!” you might say, “There’s the deflation!” Wrong. Why? Those expenditures are counted in economic statistics. They are already included. The only potential miscounted value here is the qualitative improvement of tablet + smartphone over laptop + dumphone. As per above, there’s plenty of reason to think the BEA gnomes suck at estimating this qualitative improvement, but they do try. You have to do a lot more than say “c’mon man” to convince me that this is driving a systematic mismeasurement of inflation or productivity. A sense of scale is important here. $450ish a year in expenditures in comparison with a per-capita GDP around $51,000. Arguing the mismeasurement explains to a 1% CAGR gap for 10 years is to say that the mismeasurement of value is larger in magnitude than the expenditure itself and has compounded by that amount for an entire decade. I think I’m the one who gets to say “c’mon man” here.
19. June 2015 at 01:37
@ssumner – I do want to believe you, really. I honestly want to see your theory work. So do consider a future post on this:
Sumner: Ray, You said:
“Show me when the Fed jump-started the economy with an exogenous interest rate change”
If only you knew how funny that was.
And I have entire posts on how MM is falsifiable. Have you forgotten Krugman’s 2013 “test of MM?”
How funny is it? Mystery writing is your forte, Dr. Scott “Speaking in Code” Sumner.
Krugman’s 2013 test of MM is what you consider as evidence of MM? Is that where he said 2013 sequestering austerity would be a disaster for the USA or is it another point? In any event, Krugman believes in MM. He just does not think it is better than fiscal policy. In fact, he called for adoption of your NGDPLT a while ago. This is too funny: you are citing somebody who believes in MM as evidence that MM is falsifiable and as evidence it is correct? Please provide me a link of a paper that shows 95% confidence limit stats that show a cause-and-effect between Fed action and the economy’s reaction. If you do that, I’ll go away and sing high praises to you forever.
19. June 2015 at 04:18
Matt,
“The only potential miscounted value here is the qualitative improvement of tablet + smartphone over laptop + dumphone.”
Wildly incorrect. It’s not “qualitative improvement” it’s real CONSUMPTION actually measured in how much to not live back then?
Uber requires smartphone, so that means that all the new cheap rides consumed by people – that fits under your “qualitative Improvement”
But then there’s instagram, and pandora, and netflix, and and and – and taken together they generate a consumption lifestyle that DWARFS the consumption of 2005. It becomes virtually impossible to imagine living back then. Watch an old episode of Seinfeld…. I submit to you that most people would rather go live in a third world country with bandwidth & a smartphone, rather than live in the first world Manhattan past with pay phones and desktops.
As I said to Scott, takes best 7 years in American History of consumption improvements – name them X. 2008 till today is X+Y.
So whatever the official consumption gains are for last 7 years they are wrong, unless they show a real positive Y.
If they don’t even reach near X, the measurement system itself is likely to be meaningless.
My point Matt, is that it doesn’t matter is wages fall, or “poverty” rises, or any other silly metric, ALL that matters is consumption, measured by asking, would you pay to not live back then.
I know this sucks to hear, but if your model doesn’t end up with people choosing to get into a VR pod and never getting out again BY CHOICE, your model won’t be able to explain what is going on.
19. June 2015 at 05:21
Ray I will explain it like this. If poor people really would “rather have a cell phone than indoor plumbing, all things being equal” that means that poor people are spending more on cellphones and less on toilets. So the market is therefore pricing their preference accurately. So GDP measurements are accurate after all.
To me someone choosing a cellphone over a toilet is not surprising or even note worthy. Some people prefer to drink alcohol to eating dinner, all things being equal. Does that mean alcohol isn’t being priced accurately? I see most people on their cell phones watching cat videos, playing candy crush, and I hear them arguing with lovers or creditors. I don’t really see the missing GDP on this.
Cell phones are a great technology and so is the internet. But the evidence that it isn’t being priced accordingly is lacking. Another example: I know many cord cutters. They can’t afford cable TV and cable internet. So they get rid of TV. The Internet is far more valuable than TV. But the internet isn’t free, and for some it isn’t cheap either. consuming more of one thing and less of another seems like standard economics of pricing to me. See what I mean?
19. June 2015 at 07:51
Benny, what you miss is that the “declining wages” and “economic inequality” etc you moan about – these are direct effects of DIGITAL. The movement to digital literally CAUSES the thing you hate. What’s more, the great thing, everybody wants DIGITAL STUFF is so worth it, you not being happy (having inequality) is worth it.
The point of this whole discussion is to make sure the decision about the economy are left to the tech guys more so and more so and more so….
By not allowing GDP to be the measure, we crowd out the academic Econ / bureaucrats – and leave the status and decision making to the Libertarian technologists.
19. June 2015 at 07:58
@Benny– I see what you mean. I think this WordPress forum is not conducive to threads and therefore Morgan and I seem to be talking past you. Your concern is with hedonic GDP, as per https://en.wikipedia.org/wiki/Hedonic_index
19. June 2015 at 08:07
Morgan-
You clearly are not understanding what I am saying, both on explaining the difference between 1995-2005 and 2005-2015 and on what I mean when I distinguish qualitative vs quantitative.
First, funny you mention Seinfeld. Seinfeld ran from 1990-1998. It is not about the U.S. of 2005, it’s about the U.S. of 1995. We agree that consumption quality now vs Seinfeld world is vastly better. My contention, and what the productivity numbers say, is that of the total improvement 1995-2015, the rate at which improvement happened was higher before 2005 than since.
Next, Uber. That Uber is (usually) cheaper than the cabs it replaces is a quantitative improvement, in that the price of getting from here to there is lower. No one is suggesting this improvement does not exist. The argument from skeptics like me is that because there is an observable transaction with an observable price change involved, that improvement is being captured by macro statistics. It’s not missing, it’s alredy there; “All those cheap rides” are being counted. The improvement that probably isn’t getting captured is the qualitative one, which is that the overall experience of an Uber ride is better than a cab. It will usually pick you up 5 minutes sooner, it wil be a recent model Toyota rather than a 1997 Crown Vic, it won’t have a whiff of drunken vomit, and the driver will have command of the english language. These are the qualitative improvements. Would I be willing to pay 10% more for an Uber ride over a cab ride? Sure. Would I pay 2x for it? No. Maybe you would, and that’s why we disagree over the scale of the qualitative improvement. As a matter of personal taste I usually don’t place a high value on intangible service quality. When it’s 1 am at the bar and Uber has 3x surge pricing, I actually call the cab company instead.
Next stop, Music streaming. Spotify hit 15m paying users / 60m total this year. In total, there are probably something like 30-40m paying music streamers and maybe 120m total. The US population is 320m, of whom 80% (256m) are 15 are older. By simple math, at least half of the US can easily imagine life without streaming music because that happens to be their present lifestyle. Furthermore, revealed preference says even people who do use it don’t place life-altering value on it. 75% of music streamers are aware of the option to pay $10/month or so for a non-ad, better selection version and say thanks but no thanks. Presumably the half of the population that doesn’t even stream for free doesn’t place all that high a value on it either. Furthermore, the value-add in music streaming is rendered highly questionable by the fact that none of the companies in the business make any money. On the other hand, Apple has absolutely crushed it selling music through the iTunes store. As a qualitative matter, streaming music has some advantages over buy-and-store-on-device as a delivery mechanism, but it’s not really all that much. You know what definitely existed in 2005? Buy-and-store-on-device as a delivery mechanism. The number of people whose craving for variety is so large as to make storage capacity a limiting constraint is miniscule. Music is also very age-limited consumption good. For most people the consumption value of music (and especially of music variey) decays pretty rapidly by age 35-40. The median US age is 37 and going up.
On your “pick 7 years” challenge, I will accept and choose the mid-1870s as my horse. I’ll have to go and dig up Friedman’s monetary history for exact start/end dates and figures, but he and other economic historians estimate that real incomes & consumption approximately doubled over a 7-8 year period then. In the whole of human history the only comparable event is the growth of China in the 2000s. It was the time when railroad building and industrialization exploded after the dislocations of the civil war had settled. If you read a standard US history book it will describe the time as “troubled” economically speaking because there were 2 nominal recessions. I say “nominal” because both fiscal and monetary policy were incredibly contractionary, as the policy goals were to deflate the dollar by apprx 60% to get back to pre-war metallic value and to simultaneously retire the debt incurred during the war. Nonetheless, even despite austerity on a scale unimaginable to a denizen of the 21st century, real consumption grew at essentially a 10% CAGR.
19. June 2015 at 10:28
Ray, I suppose you are right that these comments are not the best form of communication. I never read anything Morgan writes and I don’t always agree with Scott but I have been thinking more about GDP measurements, inflation, and quality of life lately.
19. June 2015 at 18:56
Ray, You said:
“95% confidence limit stats”
There’s your problem right there. Never be impressed with 95% confidence stats.
I won the first 12 games of blackjack I ever played. That’s like 99.99%. Do you think I’m good at blackjack.
20. June 2015 at 09:36
Matt,
“Next, Uber. That Uber is (usually) cheaper than the cabs it replaces is a quantitative improvement, in that the price of getting from here to there is lower. No one is suggesting this improvement does not exist. The argument from skeptics like me is that because there is an observable transaction with an observable price change involved, that improvement is being captured by macro statistics.”
INCORRECT. And I think this gets to the very base root of my notion of digital deflation…
You are observing PRICE CHANGE and I’m observing CONSUMPTION.
Imagine a luxury good, a taxi ride.
Now drive that price down. Guess what WHO CARES? (Never reason from a price change -lol)
What I’m measuring is RIDES (consumption).
Now if you look at the human population, from the .1% to the completely incapable bottom 10% (I’m speaking of “lifetime incapable,” not he bottom 10% this year), and you visualize Tax RIDES as a luxury good, I measure the justness of our society on two axis:
W: Rate of creation of new luxury goods. Higher the better.
Y: Rate at which new luxury goods reach poor. Shorter the better.
This captures the qualitative and quantitative decision making about our economy thru the Digital Deflation lens. Because it focuses on Do the poor get taxis? How many? Instead of Do the poor get a higher wage? Because a higher wage, doesn’t tell us about HOW MANY TAXI RIDES?
Now if and ever growing percentage of Luxury Goods are software! Well.. then we get them distributed faster AND we want to FAVOR more software based luxury goods.
I’m willing to apply our metric to goods – think about luxury purses and their knock offs… but let’s do software first.
20. June 2015 at 10:01
Matt,
Music: a good friend of mine is Mike Robertson, another was VP at Pandora who helped the first iPod,I knew Janus and Niklas through-out Kazaa, I even advised RIAA. I know the numbers 🙂 That streaming radio is just radio and shouldn’t pay licensing fees is a mistake. But let’s not do that, let’s instead open up our eyes to global (and US) CONSUMPTION of digital music… How many songs consumed, How wide a choice, How many of the bottom half on globe are getting it, How can we speed this process up?
Matt, a good way to measure music consumption after observing a massive increase of digital consumption of songs is to measure consumption of tickets and shows – if you note concert going is increasing AND concert ticket prices are going up, we can assume this is bc there’s a lower barrier to entry when consumers get to consume more – quantitative and qualitative (choice) – music.
It’s like I show you a giant increase in consumption of free music, then I show you more concerts and higher ticket prices (which portends more concerts)… and instead of notice a massive increase in consumption, you pull out some graphs and say, “that’s not what my model says”
2005 was a yawner. All the good stuff happened 2008+ when the atomic economy crashed.
20. June 2015 at 14:09
The price of light over 700 years:
https://twitter.com/MaxCRoser/status/612309197062021120