- Date published:
11:41 am, March 11th, 2013 - 35 comments
Categories: Economy, internet - Tags: economist
Reading the Economist this morning and saw the pseudonymous Free Exchange blog post on the economics of internet – specifically “How to quantify the gains that the internet has brought to consumers”. It was followed up by a post by Hal Varian, the chief economist at Google. I figure that I’d write about where I see the true economic value of the net because they seem a bit limited in what they’re writing about.
The difficulty that economists have with structures like the net is that they seldom look at why it was built and how it is sustained. “Free Exchange” gave a excellent example of the high-end consumer use of the net.
WHEN her two-year-old daughter was diagnosed with cancer in 1992, Judy Mollica spent hours in a nearby medical library in south Florida, combing through journals for information about her child’s condition. Upon seeing an unfamiliar term she would stop and hunt down its meaning elsewhere in the library. It was, she says, like “walking in the dark”. Her daughter recovered but in 2005 was diagnosed with a different form of cancer. This time, Ms Mollica was able to stay by her side. She could read articles online, instantly look up medical and scientific terms on Wikipedia, and then follow footnotes to new sources. She could converse with her daughter’s specialists like a fellow doctor. Wikipedia, she says, not only saved her time but gave her a greater sense of control. “You can’t put a price on that.”
The reason that the internet was so good at doing this task is because that was exactly what it was designed to do. As a bleeding edge user of the nets over the years, that is how I’ve used it.
In the mid-late 80’s, I was in Dunedin and picking up unacceptably high international toll bills as I regularly logged on to BIX in the US. I even picked up some usenet feeds using 2400 baud modems at a few kilobytes per second. It was a lot easier and cheaper searching a text feed than trying to dig through the pile of manuals, books or old computer magazines that piled up against the side of my desks (or indeed buying them). And I was often after material that had never ever been published.
By the early 90’s I used the newly arrived local usenet feed which I took megabytes ( 🙂 ) of and gopher. By the mid-90’s I was using Netscape and Altavista and the world wide web that we still use.
How I was using the net and the search engines was for an exclusively economic reason. I was writing code and maintaining systems for various computer related jobs, while at the same time hauling myself up the skill set chain in a rapidly changing environment. The economic values of that are literally incalculable because not only was I shifting professions from management to computing (I was in Dunedin doing an MBA) without bothering to spend money on formal training, but I have since used it to stay relevant in a profession that burns out people through skills obsolescence at an inordinate rate.
It is the same today. I just spent the last few years building a new product with a number of programmers, electrical and production engineers. All of us would probably have spent time during every working day using the net via the web or email digging out the bits of information that were required to put this product together. Components of the system were literally coming from all over the world to a final assembly locally. Simply organising the supply chain would have been a major operation a few decades ago without the net. These days it is simpler and *lot* faster than I remember from my days as production manager working through sluggish supply chains. Whilst that is an economic effect that may be measurable, I’d hate to have figure out how to get the data.
As usual I was working in a new environment using new or enhanced tools. This time a embedded debian linux/gcc using an Arm CPU using boost/Qt4 to do a colour touch screen. This was after a year on windows/visual studio using boost/Qt3 as its tool kit helping build a system designed to span the internet and that rendered graphics using directX to screen or to a browser page. That was preceded by a few years building an embedded linux/gcc system using direct X calls for a gui for an eftpos terminal with secure keypad.
There is a awful lot of learning in those various technologies. Quite simply shifting skill sets that far and fast before the internet and delivering product would have nearly impossible. I call this the bootstrapping effect. Once you have a reasonably good base in an area of knowledge, then the resources available on the net allow you to shift areas of expertise rapidly.
Needless to say the internet is my additional brain. Most of my skillset these days (like many in the IT industry) is the knowledge to know what I’m looking for and the skill to sort through the chaff until I locate it. It is only rarely that I have to dip into manuals, and even those are generally standards that should be online.
Most of my search queries is spent simply using net as a reference. The detail of the boost and Qt libraries for me operates like this because remembering the detail of parameters on one of the many methods on one of thousands of classes that I routinely use is something that I often don’t bother retaining. If you just remember roughly what you searching for then you can get it in seconds.
This is essentially what Hal Varian was talking about as one measure of the value of the internet.
So one way to measure the value of online search would be to measure how much time it saves us compared to methods we used in the bad old days before Google. Based on a random sample of Google queries, the UM researchers found that answering them using the library took about 22 minutes while answering them using Google took 7 minutes. Overall, Google saved 15 minutes of time. (This calculation ignores the cost of actually going to the library, which in some cases was quite substantial. The UM authors also looked at questions posed to reference librarians as well and got a similar estimate of time saved.)
I attempted to convert this time to dollar savings using the average wage and came up with about $500 per adult worker per year. This may seem like a lot, but it works out to just $1.37 a day. I would guess that most readers of this blog get $1.37 worth of value per day out of their search engine use.
But the really productive use is not looking up references and I don’t spend much time on it. I expend the time on innovation and avoiding blind paths. Finding out how other people have previously attacked a type of problem and the pitfalls and solutions they found is amazingly productive. These problems could be simple or very complex, but they usually involve a developer or engineer spending weeks doing experimental development trying to define what the problem is so they can look for a solution.
Most of the time in a global network, you’re not the only one who has the problem. The others with the problem are found in one of the innumerable blogs and question sites1 where people have documented the process that they and others followed when looked at the same or similar issues. Even if I didn’t find a solution on the net after a hour of searching, then it usually directed my experimental programming to areas that still offered a hope of a solution. Sometimes whatever I was looking for would not show up at all, which itself was highly useful information. Anything you find (or don’t find) is pure gold because experimental programming can often take weeks.
So this is just my work life. I won’t even mention the effects as political blogger, my personal and family life, or just how I entertain myself these days. They’re somewhat larger.
Economics has very little hope of being able to analyse the value that the internet has in our modern economies. In particular into the innovations that drive our modern economies changes and growth. It is ubiquitously embedded in most businesses these days to a degree that would have seemed fantastic even a few decades ago. While I wish economists well in their continuous attempts at measuring the effect of the internet, I think that it will be a futile endeavour.
1. My especial thanks over this last project go to Stack Overflow and Linux Questions, who probably answered half of my questions either directly or from links in their sites.
That old joke ‘if you placed all the economists in the world end to end they still wouldn’t reach a conclusion’ comes to mind.
For every person who gets helped to understand medical treatment with correct information, how many pick up a bunch of inaccurate woo and decide to self medicate their brain tumour with homeopathy?
That is always a problem. But it always was. Have you ever read any books on 19th century home medicine? Or for that matter what the doctors used? I have in one of my more morbid periods of historical reading and they would make your hair curl (and probably fall out).
When reading *any* material you have to approach it with a degree of sceptical cross referencing. Of course in my field of work interest, this really isn’t a problem because the people who know what they’re talking about give quite clear directions on what they’re looking at at and why. The ones who do not will usually manage to make that quite clear within a very short period of time. They tend to read like people hooked on the IRC.
But I’m with Larry Niven on this one. “Think of it as evolution in action”
Computers have the fundamental advantage of being mostly unsusceptible to woo in a manner that can be easily demonstrated. Holistic debugging via homeopathic correction can easily be found not to work.
(On the other hand, there is the (cyclic) belief that untyped languages are more efficient. And then there’s management faddery, like standup meetings).
Nice technical history LPrent: over the same period of time I have worked in IT and telecommunications as-well.
I remember years ago replacing 2400 baud modems that supported 4 stenographers adding records from the Courts in Chch to the Wanganui computer. We revisited 4 months later and 2 were redundant as the 9600 baud modem had cut down character input times.
Then came the ATM machines…lots of bank tellers disappeared overnight as the wage bill got slashed by automation.
Technology is really good at removing human functions. The only economic “value” I see in IT is the ability to remove workers functions, and the internet facilitates this substantially. Fortunately for workers the return is diminishing and the costs of supporting the increasing overlays of technology are building up.
I think you got a word quite wrong there. Technology is really good at removing inhuman functions. Machines are only really good at doing dumb, boring, and above all insanely repetitive tasks. Humans are good at handling new situations, dealing with other people, and generally learning. What they are not good at doing is doing anything as repetitively as a machine can. And machines are far more stupid than any human I have run across.
Most people when stuck as a old-style pre-ATM bank teller or process worker (I’m more familiar with the latter) doing something that any old machine could do would tend to spend their working life hanging out for the next break or heading home so that they can spend time relieving the boredom dealing with other people. They did those jobs for the pay packet and bugger all else.
Having done quite a few of those types of jobs myself and worked with people doing them I’m rather happy that they’re disappearing. If you want to drive guinea pigs insane all you have to do is to put them in similar situations. Treadmills leading nowhere.
But economies change. We used to need to have people doing those tasks as a kind of robot. These days we do not because we have actual robots. And in this country and in most we have been slowly sucking up just about every adult human available to help run (more or less intelligently) an more complex society with more people in it.
Consider that when I was born in 1959, the country had 2,359,700 people and a workforce that was somewhere about 900k-1000k people (because most women still didn’t do paid work). The country now has ~4,460,088 people and workforce of something like ~2.3 million despite a rapidly increasing number of retired people.
So despite the technological changes you’re talking about both the numbers employed and the percentages employed actually increased and did so quite markedly. Why? Because unlike machines, people are flexible and changed what they worked at. These days it is the expectation. Kids (ie anyone younger than 30 😈 ) today act with incredulity when I tell them that I once did a development job for 11 years, that my father worked at a single workplace for 20 years, and that my grandfather did the same for over 40 years
But removing the inhuman jobs from humans doesn’t work for all – from the teara link
And that was at a period when we were creeping as close as we have come to come to full employment in the last 25 years. Humans have to learn to how to learn. Getting them to have the opportunity to do so is just about the most important thing we should be doing.
And that’s something that National Standards and everything else this government is doing to the education sector won’t do. Instead it will do this but for the children and there won’t even be pay packets:
Really not what you want for an education system.
Lprent, I like your “inhuman” point: yes the functions could be fairly dire. Conversely they did involve me as the customer in human to human contact, dumb machines cant do much else.
My problem with the whole work displacement was the economic theory of freeing up capital to generate new jobs elsewhere….hmmmm. Or the other hoary chestnut of jobs able to be offshored across the net. Nuff said.
Ennui – you raise a vaid point.
As much as Lprent is right in arguing that many “inhuman” tasks have been replaced by computers performing repetitive and highly complex functions in a fraction of time, we have now a society where computerised systems are in high use, and where the human contact is being abolished to a degree that it leads to alienation and potential for marginalisation of whole groups of people, who may never feel that comfortable using certain technologies.
Look at the online services in use by government departments and also companies offering services.
It is immensely hard to get heard and served in some cases, be this by WINZ staff or whosoever, if one is not that IT savvy and online to check on things and to arrange details.
There will be the day, where going to their offices will not get you anywhere, unless prior contact was made by email or online. It is just all a beginning what we have. Also look at the phone systems, where overloaded call centres leave people waiting for up to an hour or more, getting nowhere.
The list could go on.
The world is becoming more inhumane in that sense also, due to computers and the internet taking over so many contact and interactive functions.
But is that a function of the technology or the fact that we’re working longer and longer at make work tasks (BTW, I put call centres into category) and being paid less and less?
Xtacys fears are (in my opinion) validated by the use of technology to map process/ procedures within boundaries that proscribe responses that are outside of what the organisation handling it wants to happen. Humans must by no means be allowed to have capacity to decide, person to person, automated systems help faceless entities like corporates avoid scrutiny on individual cases. Very dangerous, very f*****st..
Which is why Work and Income need to change their policy and consider internet access an essential need and include provision for this. As far as I can see this would be a good thing for cities and towns, less stress on clients and staff and the use of communication tools such as Skype (useful for lipreading, those who have literacy / comprehension issues.)
For those in rural areas where internet access could cost thousands so will be problematic without a service centre close by, but more tech should mean lower costs if properly implimented.
BTW has anyone tried to test Firesheep in a work and income service centre yet? Pretty sure it’s all wired, and secure. Roll out date for new kiosks is May.
Technology’s good. It has the potential to do all sorts of fantastic and freeing things for people. The problem is that those who control it don’t allow everyone to experience the benefits. Technology works to reduce labour costs, but the 40 hour working week is still what’s expected, on a numner of different levels, from most “workers”. The concept of work and leisure has only changed positively for those who control the technology.
That’s exactly what we’ve been seeing and it really is doing our society harm as poverty increases because of it. But that’s not technology doing that but the dictators in control of the corporations.
And work usually expands to take up available time rather than absorbing new workers
“hauling myself up the skill set chain”
…..erk! I almost gave up there Lyn. Thankfully I didn’t – but there goes a declaration of my prejudice.
But CHRIST devoted to son ALMIGHTY ….. how sick am I of hearing about skUll shortages and the like as the corporate-wedded (human RESOURCE) agencies look to fill positions as many [obviously I can’t name] lie in fallow wasteland. And now even IF they might, pardon they if the result was a two fingure salute.
No, I’m quite happy to sit back and watch as wheels are re-invented; egos are stroked; and as history is repeated.
And when I do, in I’m incomplete, UTTER amazement. I mean ,,,,,my own son could probably gave offered a solution (or his mates) to your latest dilema ( I’m naturally enough DISINTERESTED), we’ll jump ahead an inch or two.
My point above is that there are actually several 50+ yo’s who’ll never (and I mean EVER) get a look in.
Anyway … best of British and all that kaka – ka-ka-ka-ka-Hah.
I now can’t remember what level of ‘reply’ I’m installed in, but a funny thing happened on my way back from the Wairarapa yesterday.
Turns out that one of the hUtch-Hoikers I picked up was the son of a certain Sth American Ambassador (currenlty touring with a slUpppery DUck).
No better a political conversation could have been had – since that ambassador was utterly devoted to son and his various interactons.
SHORT MESSAGE if our glorious Master thinks (in ANY way) that the snub of Hu-GO’s funeral, the tittering and twinkering of Johny’s & BronUZZ’s little little jaunt – alongside a complete and utter fuckwit that EVEN JK himself had to apologise for with his interaction with an electric fence – then they must think we’re really really stupid and about to engage with JK on the basis of his presence.
Well ….. GOOOD!
Personally, I felt the need to apologise for the manner in which NZ has a fool (a semi-literate, self-serving one at that, egotist, money (and anything) trader leading the country).
Hey LP, the editing, the backspacing et al did make the above read quite well. What’s submitted is NOT was originally intended.
But since I got the “Duplicate comment detected; it looks as though you’ve already said that!” comment, alongside some pretty pathetic response times elsewhere, ANYTHING could happen (the the next half hour).
Quite OBVIOUSLY – most of my corrections to the above diatribe did not take, and quite a big proportion of what I thought I’d submitted looks nothing like what appears. (NO I’m not drunk)
Still working on the broken re-edit. In fact I should be doing it now rather than writing relies…
But hey, I’m on an actual holiday slothing around and with zero family present…
Yeah. I love the internet, and couldn’t live without out, but I don’t hold to the ever increasing internet theory. The internet has a physical footprint, and physical infrastructure that underpins it, much of it in either benign or malign neglect. It’s energy requirements, whilst small compared with say transportation, are still pretty substantial (the old 1/4 of a kettle of water per google search click is a commonly quoted statistic, although getting reliable information on this is next to impossible).
What I guess I’m interested in is the long term sustainability of the internet, with full lifecycle replacement costs. I can’t see it lasting in its current international form forever, and we’ll probably begin to see the decline of it within our lifespans, as unpopular as this may seem.
Hooked into the same old problem of cheap energy.
But I’d point out two things related to that. If you ignore the PC’s at the end of the chain. Much of the structure of the net is held in rather large data-centres scattered around the globe. They chew a lot of power but are also non-mobile plants that are just aching to be optimized for their energy and equipment footprints as well as economies of scale. A trend that has long since been going on and is rapidly increasing. The same applies to transmission systems. There will be cost shifts in those, but I suspect they will be downwards rather than upwards
The other end, the last mile is where the cost problems will kick in.
But the problem of costs only really applies to the low economic value parts of the net – which I didn’t talk about above. The internet of trivialities has a future problem, but I suspect that the uses that I mainly use the net for do not. Most businesses using the international internet get a return that is far in excessive of almost any possible cost.
I guess the difference in view that I see a different net to the one you see.
Interesting point Lynne. The internet I use mostly is the “engineering net”, as in the collection of largely pre-WWW open source technologies that underpin its communication function. That will keep going for an indefinite time, if not on wires or fibres, then maybe on the airwaves (at a massively reduced bandwidth). These technologies have huge utility beyond that related to bandwidth availability.
The internet of trivialities is another matter of course. This needs international bandwidth, and big data centres. I see the advertising model that sustains this collapsing before the infrastructure does. It’s already happening of course – and when it does, these services will go back to largely different subnets, based on a subscriber model. You’d have to choose the Google net, or the Microsoft net, or Apple etc…
I guess, at some point, the cost of keeping high capacity international links in service may exceed the benefit, and we’ll be back to a nationwide internets, largely running thin-client public services.
And once we get start getting fibre out to the last mile that will start dropping as well. No more millions of kilometres of 50v line running everywhere.
Not on a 2400 you weren’t. Most I ever saw was ~225 bytes per second.
That’s because modern economics works on competition and restricting access to that knowledge so that it can be charged for rather than people across working together and sharing that information. The latter way produces far more value than the way that economists expect people to work but it can’t be charged for and so it can’t be measured in a monetary way (ie, there’s no profit).
You are so right. 2400/9 = 266.67 bytes per sec. Frigging hell…. Umm.. You have to have a 9600 baud to get to 1kB/sec.
I guess I was remembering the 14400 zyxel that I used for quite some time in the 1990s. It was a while ago.
Your other point about the value of the internet is the way I view it as well. It has a utility value to the rest of the economies that is far in excess of its costs. The nearest analogy that I can think of to it economically is the successive rises of canal networks, railways, and roads which had a similar effect. But with those the effect was gradual enough to be measured. The net kind of gushes into new areas so fast that half the time you find out afterwards.
Also, I still believe the commercial economics of the internet are similar to aviation. Since airliners were invented, they’ve got shiner, mostly faster (though we peaked in 1972) and more economical.
Shareholders in airlines and aircraft builders have generally lost money, often in wads. They only keep going because so many people love the smell of the kerosene.
The internet’s exactly the same. Examples of huge fortunes being made (Howard Hughes / Mark Zuckerberg) are just the exception that proves the rule.
But that is my point. Who cares about the odd fortune at the frothing edge of consumerism. It is mostly speculation anyway.
What matters is the way that it is changing the nature of all business and resource movements merely by being there and being available. I can remember dealing with companies in India by letter in 1981 with a turnaround time of weeks. Or the telephone tag with trying to organise transporting bricks from Kamo to the Bluff. It isn’t like that any more and 90% of the reason is because the net allows for better async communications.
Individuals have access to more data than at any other time in human history, but comparatively less access to sorted information.
just a Tool (with discernment required imo)
On this topic, the question posed and attempted to be answered is a bit like: What is the economic benefit of being able to read and write.
In short, it may be hard to measure, but it is immense, and it increases or declines depending on the level of skills and qualifications obtained, and the degree any technology, even as simple as “written language” or “script” is developed and applied.
It is an interesting subject for sure.
To me the internet is a new means of communication, revolutionary even more as the invention of the telephone once was. It is a means to an end, and it is in constant development, so few if anybody can predict, where it will all lead to.
My concern is, and I stated it before, that given the limits of accessibility due to “literacy”, knowledge and education, ability to afford, commercial and non-commercial controls, state and international sanctions, it is a danger there to become the technological communication network that may be there for a privileged few only. What is offered via the web can be manipulated, as much information is indeed not reliable, subjective, trivial, personally chosen by biased, self-interested or ill-intending individual persons or particular commercial or non-commercial organisations, thus requiring the user to develop skills to discern between worthiness and unworthiness, truthfulness and untruthfulness and so forth.
A constant involvement, continual learning and exchange with others, awareness and also decisiveness to defend privacy and freedom of use is needed, otherwise the web as we know it now will soon be one of the past.
I have often thought on the questions raised here. What I take away from all this automation and IT is that it will need a large societal infrastructure to support it. Higher education will become almost universal, and obtainable by all. Tens of thousands of instructors and caregivers, all the way from primary to tertiary needed to create and nurture the specialist technicians able to construct and maintain the whole edifice. (both hardware and software). Class sizes must drop, requiring many more teachers. Education will become a huge industry in itself. Maybe the biggest industry of all. And not all of it technical. I envisage a huge flowering of the arts as well as the sciences. After all, we do not live by bread alone. Possibly a lot of this education will be done on line. Mooc for instance will play a bigger and bigger role. Very small class sizes with instructor and students with fully interactive high speed internet, may be the possible future of education.
However we are in a race. A race to create a fully sustainable internet that can survive independently of the collapse of global society as we know it.
I wonder is anyone actually looking into this?
My attempts to weave a terminal out of wickerwork has not met with much success, so far.
instead of the economic value of internet,
the economists should be evaluating,
the economic value of interest,
Like the economic “value” of water, which is clarified in times of drought, the economic “value” of the internet will quickly become apparent when it breaks down, global electrical mayhem, chopped cables etc. Economics attempts to descibe/model the allocation of scarce resouces. Welfare economics -where the stricken farmers are now – considers “value” in monetary terms – for everyone there is a social cost – including the poor cows. We can do without the internet, but not water. There is no such thing as a “fully sustainable internet” especially when storage facilities rely on close water sources for cooling.
This is another example of the net producing far more value:
What I see here is the beginning of the end of the consumer society. Instead of everyone owning their own major consumer good they can get together easily and share. This will result in less resources actually being used and so, under normal capitalism, a decrease in the economy but we would also see an increase in social well being.