The Tragically Neglected Economics of Abundance

I'm preparing for my talk on Long Tail economics at O'Reilly's Emerging Technology conference in ten days, and I've run into a slight problem. The Long Tail is all about abundance: the economic effects of infinite shelf space. Unfortunately, neoclassical economics has virtually nothing to say about abundance. Indeed, the economics of abundance is almost exclusively the domain of extropians, a few other transhumanists, and science fiction writers. How can this be?

Well, for starters the classic definition of economics is "the science of choice under scarcity". That's a warning sign right there. From Adam Smith on, economics has focused almost exclusively on behavior within constraints. My college textbook, Gregory Mankiw's otherwise excellent Principles of Economics, doesn't mention the word abundance.  And for good reason: if  you let the scarcity term in most economic equations go to nothing, you get all sorts of divide-by-zero problems. They basically blow up.

But clearly abundance (AKA "plentitude") is all around us, especially in technology. Moore's Law is a classic example. What Carver Mead recognized in 1970 when he encouraged his students to "waste transistors" was that transistors were becoming abundant, which is to say effectively free. The shift in thinking from making the most of scarce computing resources to "wasting" cycles by, say, drawing windows and icons on the screen led to the Mac and the personal computing revolution. To say nothing of the scandalous profligacy--a supercomputer used for fun!--of a Playstation 2.

We also have similar abundance laws working in storage and bandwidth and virtually everything else digital. Outside of technology, the green revolution brought abundance to much of agriculture (so now, to prop up prices, we pay farmers not to plant their crops).  And what is the motive force behind China and India's rise if it is not abundant labor, allowing them to, in a sense, waste people?

Even ideas can on some level be considered abundant, because they can propagate without limit due to their "non-rivalrous" nature (a feature they share with fire and digital media). As Thomas Jefferson, the father of the US patent system, put it, "He who receives an idea from me, receives instruction himself without lessening mine; as he who lights his taper at mine, receives light without darkening me."

More than a decade ago George Gilder, the apostle of abundance, offered a good way to think about all this. In an interview (in Wired, as it happens) he said:

"In every industrial revolution, some key factor of production is drastically reduced in cost. Relative to the previous cost to achieve that function, the new factor is virtually free. Physical force in the industrial revolution became virtually free compared to its expense when it derived from animal muscle power and human muscle power. Suddenly you could do things you could not afford to do before. You could make a factory work 24 hours a day churning out products in a way that was just incomprehensible before the industrial era. It really did mean that physical force became virtually free in a sense. The whole economy had to reorganize itself to exploit this physical force. You had to "waste" the power of the steam engine and its derivatives in order to prevail, whether in war or in peace."

That suggests a way to put this in an economic context. If the abundant resources are just one factor in a system otherwise constrained by scarcity, they may not challenge the economic orthodoxy. They are then like learning curves and minimized transaction costs--drivers of production effeciency that serve to lower prices and increase productivity but do not upend the economy as a whole.

And, indeed, the abundance of the Long Tail, for all its power, is surrounded by such constraints. Although there may be near infinite selection of all media, there is still a scarcity of human attention and hours in the day. Our disposable income is limited. On some level, it's still a fixed-pie game. Offer a couch potato a million TV shows and they may end up watching no more television than they did before; just different television, better suited to them.

By analogy, Moore's Law just shifted the scarcity bottlenecks from the CPU to the user. Now the limits of productivity are not instructions per second but ease of use, which is why that's where those transistors are being so wastefully deployed on the desktop today. The consequence is that our productivity increases, a phenomenon entirely compatible with economic theory. But that happens at a pace of 2-4% a year, not the annual 60-70% increase of Moore's Law. In other words, there are still limits to growth.

There's still a lot to be said about the economics of the Long Tail, from how the shape of the head changes as the tail opens up to the question of whether tails have a nested heirarchy of mini-tails. I hope to have enough data to answer those questions by Etech, or at least phrase them well enough so that I have the answers in the book. I'll certainly be further along than I am today; I've got some interviews with good economists next week, and they're sure to be of help.

In the meantime, it's worth noting that the fact that economics doesn't have a good theory for something doesn't mean that it isn't real. For instance, economics is still struggling to explain growth itself. Seriously.

Comments

[Tom Guarriello]

Chris, you should check out Grant McCracken's book, Plentitude. Also his blog: http://www.cultureby.com/trilogy/

[BillSaysThis]

The economics of abundance has been a topic of interest to me since reading G. Harry Stine's novel Manna over 20 years ago (written under the pen name Lee Correy). Moving the bottleneck is an interesting phrase in this context, but one aspect of the situation you don't much mention in this post is the coming resource contraints.

Not just petroleum, which is perhaps most obvious, but also arable land and potable water. The green revolution to which you refer has delivered a tremendous growth in farmland productivity but at a huge cost. One wonders how productive Central California farms will be in, say, 20 years unless some relief is found from current techniques; a "lot of farmland in California is close to being ruined from over-irrigation; you can see the salt precipitates in the fields off Interstate Five in the Central Valley today," to quote Jim Kunstler.

[Robert Paterson]

Chris
Are you familiar with the work of Paul Romer and of Td Homer Dixon - a big help in looking at abundance?

[Greg Linden]

I think you're confusing abundance of choice with abundance of resources.

The long tail is about abundance of choice. As it becomes easier to get information about small providers of products and services, consumers discover more choices, choices they didn't know they had before.

The long tail isn't about abundance of resources. Merchants in the long tail consume the same scarce resources that large merchants consume. Resources are just as scarce as they were before.

An interesting question here might be whether reducing the cost of information, making it easier to find products in the long tail, increases efficiency and productivity. But I don't think that the long tail challenges the basic principles of economic theory.

[Chris]

Greg,

I disagree. I think the Long Tail is about the abundance of choice *provided by* the abundance of resources. The example of virtually-free storage, which is the source of the "infinite shelf space" that we see everywhere from iTunes to the iPod, is a case of abudant resources. Likewise, virtually-free (ie, abundant) bandwidth will be the driver of the coming infinite channels of IPTV.

In the case of the iPod, it required Steve Jobs switching from scarcity thinking ("how many songs can anyone really listen to on single outing, anyway?") to abundance thinking ("why not carry everything everywhere so you don't have to choose ahead of time?"). That's the sort of "waste storage space" mentality that's at the core of what I'm trying to express in economic terms.

[Scott Johnson]

Something you might want to check out:

"The Economics of Abundance" by Friedrich Hayek in The Critics of Keynesian Economics, Henry Hazlitt, ed., New York: New York University Press, 1983, pp. 125-130.

[Martin Weber]

Bingo! Economics is centered around the principle of scarcity. So how does it deal with something that does not fit into the equation? Impose scarcity upon it.

While many resources became plentyful only in the last few decades there has always been one resource that was in infinite suply: Information - and as a derivate - knowledge. This resource is extraordinary as it grows more effective the more you use it. In opposite to resources like energy, labour etc. which get depleted through usage.

As "know-how" became increasingly important (say, the last 100 years) economics found a way to impose scarcity upon this resource: Patents. So our economy work as we are used to it and everything is fine again.

Sorry for the sarcarsm, but I (as in: in my humble opinion) think patents will eventually kill the long tail in the long run.

[Gen Kanai]

Chris, how about W. Brian Arthur's work on increasing returns? Arthur's work may be a very good lens with which to view The Long Tail?

[Dean Brooks]

Chris,

There are two very distinct traditions at work here, which only very rarely overlap in one brain. Although each has distinguished contributors, I cannot think of anyone I can cite who has written about both.

On the one hand we have the Wright-Henderson diminishing costs curve and all its siblings like Moore's Law. These seem to imply a universe in which abundance increases without limit. (One consequence being a "long tail" of diverse products whose inventory & shipping costs are sufficiently low that they can be distributed even at very low volumes for each product. Also, keep in mind the diversity is a symptom of low transaction costs, and it is not necessarily the case that the individual products are cheap or efficiently produced, as their individual volumes are not high enough for that.)

Bruce Henderson certainly thought there was no limit to diminishing costs -- several decades ago, at the start of taking Boston Consulting Group to the level of a billion-dollar business, he wrote that all markets ought naturally to tend to monopolies, as the market leader's costs continually spiral down to levels that competitors can't match. But then a few years later he conceded that market share doesn't actually do that, and that the share ratio between the market leader and its nearest competitor seems to stabilize around 2:1. If Boston Consulting Group has since figured out what causes this, they're not telling anyone for free -- but let me offer some clues.

On the other hand, there is a separate diminishing-returns curve that is of similarly wide application. Military historian Trevor Dupuy wrote about this curve extensively, showing that as armies grow in size, their effectiveness per man declines. Larger armies inflict fewer casualties per man. Larger quantities of bombs dropped in an air raid kill fewer people per tonne and destroy less property.

My own research has turned up many more applications. Supercomputers built by linking individual processors experience diminishing returns in total throughput as the number of processors increases. Fortune 500 companies that downsize due to reduced market share or whatever other cause experience *increasing* inflation-adjusted output per employee. Even epidemics experience diminishing returns -- for the same disease, infecting a much larger population tends to lead to lower average mortality rates. That includes the dreaded Ebola, which has 80 percent mortality when confined to a single small village, sees a substantial drop in mortality if it spreads to say 1,000 people.

Here is the tricky part, which Bruce Henderson at first failed to recognize. When you plot a diverse sample of diminishing-cost curves, you will see them converge on one particular logarithmic constant-slope curve. Costs tend to fall by 50 percent for each 10-fold increase in volume.

And when you plot a diverse sample of diminishing-returns curves, you will see them converge on the SAME particular constant-slope curve. Returns also fall by 50 percent for each 10-fold increase in volume. Yes, this includes the output of supercomputer arrays, or the mortality rate of epidemics.

Thus for example if I have a factory making tanks, and I send out my first 1,000 tanks into battle, they will produce a certain amount of useful output (enemies killed, ground captured) X for a certain cost Y. If I then ramp up production and manage to field 10,000 tanks, they will on average produce 5X worth of useful output, for a cost of about 5Y.

Demand curves work this way too. The first 1,000 customers you manage to find for your product will tend to be early adopters for whom it is a must-have or who simply have a lot of money to spend. As your customer base grows, you will be dealing with customers for whom the marginal value of your product is smaller and smaller. It's great that your cost keeps going down, but you don't actually get the kind of spectacular net profits that it first appears you might, because you can't sell for the price you started at.

The universe is not quite so easily fooled as the preachers of limitless abundance assume. It does not actually give you something for nothing, or more and more for the same input.

The mathematical framework that best explains all this is *non-stationarity*. As the total number of events or transactions or objects grows, the mean size keeps shifting.

... continued in next post due to size constraints ...

[Dean Brooks]

... continuing from last post ...

The event or transaction or object may in human terms be something good (money received from a sale) or something bad (cost in labor to produce an item). Some things like casualties in war manage to be both depending on whose side you are on. But from the universe's point of view, all that matters is that as the total set size grows, there are more and more small items as a proportion of the whole. The universe does not favor diminishing costs, or diminishing returns. It favors *diminishing*, period.

Of course there are many situations in which this diminishing-mean phenomenon doesn't arise. Successive generations of people aren't smaller just because the total set of people who ever lived keeps getting larger. Figuring out what conditions are necessary and sufficient for a diminishing mean to occur is fairly tricky.

Sorry for seizing your forum for a long lecture, but I've been working on this particular problem for a long time. What I just posted is a brief version of Chapter 1 of the book I'm now writing. I hope you find it helpful.

You can download several free publications with more on this subject at www.ekaros.ca. Look under "statistical publications".

Cheers, Dean

[David]

Chris, interesting post on abundance, which does seem to exist in many areas. My experience as a consumer, however, is that even with many more choices (which is a good thing), I still have the same amount of dollars to spend. While some things have gotten much cheaper, like computing power, others, like housing, have gotten exponentially more expensive. And as a producer (artist), I'm still competing for the limited attention and dollars of my potential audience/market. So even with abundance, there is still an economics of scarcity.

[Frank Ruscica]

Dean,

Good stuff in your comment.

My question to you:

What if a supplier benefits from network effects that reduce marketing costs? In this scenario, it seems conceivable that the supplier's profit margin could be preserved, or even widened, even as the price decreases...

The logical conclusion of this dynamic, of course, is a natural monopoly, in which case margins could be preserved indefinitely at some legally-sanctioned level...

Thoughts?

[Chris]

Frank,

The scenerio you describe--network effects lowering marketing costs--is exactly the proposition that Netflix and others are offering mid-list films that can't afford major theatrical release. But I really don't see how you get from that to a natural monopoly. Even a closed network like the Amazon and Netflix sites exist in the open, larger, network of the Internet as a whole.

[Dean Brooks]

Frank,

About network effects, I think that's pretty much where Amazon is right now, and Ebay, and Microsoft, and Wal-Mart . . . They get gigantic free advertising and word of mouth and other economies of scale. Yet as Chris observed, it hasn't led to a natural monopoly in any case, and is unlikely to.

I run a small publishing company so this hits very close to home. I have books that are unique, without any real competition, and I charge $200 for them. They're perfect for Amazon. I also have trade books I can't sell on Amazon because the price is set by comparable books to $50 or so, and Amazon charges small shops like me 55 percent of list price, and requires that we pay our own shipping. It isn't Amazon's economies of scale that are at fault here, but mine. The short print run of these books, and low price dictated by competition, means that they cost me too much to sell by Amazon's rules. See what I mean? It's THEIR network cost but it's still MY production cost that determines what gets sold.

Incidentally, my printer charges me on a curve that is almost exactly the x10 = 50 % discount curve I described above.

[fling93]

I don't think abundance has been reached. I think the "waste transistors" mindset is not because they are abundant, but because the costs of them are insignificant compared to other bottlenecks. But not everybody in the world (or in our country) who wants transistor-powered products can afford them, which means there is still scarcity.

Similarly, India and China's rise isn't an abundance of labor, but that they have cheaper labor than other countries. This is caused because they have more of it available than we do. Having more than someone else isn't abundance, since it's not like there isn't a lack of work to go around for that labor (they don't have "too much" labor). So the rules and equations of economics, as far as I know, still apply to China and India, as well as products using transistors.

[Frank Ruscica]

Chris and Dean,

Thanks for the replies. Re: natural monopolies, I should have clarified: I'm not claiming every case of increasing returns culminates thusly. Rather, just that it's a possible outcome, which, at first glance, seems to have the potential to preserve margin.

Enjoy,

[Britt]

Another book worth checking out is The Accursed Share by Georges Bataille

[Yuri van Geest]

Great discussion!

My 2 cents:
- I believe a distinction of many levels is necessary to sharpen the debate. The distinction between production costs and transaction costs. The production costs are lower, especially for digital content due to the falling prices of the relevant tools for production. The transaction costs are lower due to the nature of digital media.
- Distinction between old and new economies. Labor, capital, technology, energy and other natural resources versus data, information, knowledge. The latter is indeed described in the works of Hayek, Brian Arthur and Kevin Kelly. Positive feedback loops and increasing returns are not as relevant in old economies (and their restrictions) relative to new economies.
- Distinction of abundance of choice and abundance of resources (attention, money/budgets). Choice of digital content is infinite but the resources are constrained. However, even on the resource side there are tools alleviating some concerns for attention like recommendation engines (even for less popular content items) , RSS and viral marketing. And the term abundance might indeed move the economy to new bottlenecks, scarcities and restrictions, meaning the old theories might still be relevant.

Looking forward to your book, Chris.

[Atanu Dey]

I believe that the essential confusion is between the abundance of the products and the abundance of the *factors* of production. Economics deals with the real scarcity of factors of production so as to maximize the production. The end result of that exercise may be an abundance of products (and consequently an abundance of choice) but that does not argue against the basic reality that factors of production are scarce and choices need to be made in allocating scarce resources for various (often conflicting) purposes.

Information is a public good (non-rival in consumption, etc.) but the production of that public good requires private goods (rivalrous in consumption) which have competing uses. This is well within the purview of standard economic theory. So while you see an abundance of digital products which increases the set of choices, the production of these require (at least some) non-digital factors that are limited.

[Mr. Econotarian]

I think the concept that somehow "The Long Tail" breaks economics is bogus. Scarcity is a general term. Something can be common (such as iron), but still not as much as everyone would want everywhere where you would want in the quantitites that you would want. Lots of prices are coming down, but bandwidth (effective Internetworking to the end user) costs, computers cost, attention costs, and time (basically) costs.

The big economic question of "The Long Tail" is where does the tail come from? For the extreme end of the tail, there is a lack of business models, except non-profit models.

"Long Tail" distributors can make plenty of money from aggregating cheap information transactions, but most information providers will have a hard time making money.

Yes, you can download copies of "Jam on It" from Napster, but this musical work had already turned a profit back in the 80's. Money made now is just gravy. Could more "Jam on Its" effectively be created under the "Long Tail" with the questionable business model of electronic content creation? It remains to be seen.

[Kshitij Chandan]

You may have missed a point. Abundence like scarcity is a relative term. So Abundence of something is due to the scarcity of some other thing. So Abundence of supply is due to scarcity of demand. Isn't it?

Now as per the definition of "the science of choice under scarcity", the law does hold true on the manufacturing/production side i.e. meeting the scare demand, optimizing/making most of the scarce demand.

[Azeem Azhar]

This is an excellent book on the economics of abundance.

[Steve Burgess]

Another blog entry (from an essay) that references this page is here
http://www.kurzweilai.net/meme/frame.html?main=/articles/art0671.html
"The (Needed) New Economics of Abundance"

[Frank Szendzielarz]

I find it surprising that a lot of chat about the economics of abundance is not referencing the main and most influential proponents of an abundance based economy: the Technocrats. (http://www.technocracyeurope.eu/ http://www.technocracy.org) Is it only because now that global warming is becoming a fact and the abundance of digital media is an insurmountable obstacle to digital rights militants that technocratic principles are being brought once more to the fore?

This is old news. In my opinion the future economy is inevitably a mix of abundance and scarcity based systems. It is a necessity. In fact, one could say a good example of that necessity is our unintentional surplus of CO2 and how we must create an economic system that manages it.

[Hedge Fund]

Wow, great site to go along with your book, nice job. I'm trying to navigate the book publishing route myself after writing a few hundred individual articles on hedge funds.

I had actually think I have seen your book somewhere else in the past, so it was good to read a description of what your book is really about.

- Richard
http://richard-wilson.blogspot.com