Washington D.C. Public Library Case Study

Washington DC’s 37% rate of functional adult illiteracy reflects one of the most underfunded and underutilized library systems in the country. In 2004, former DC Mayor Anthony Williams launched a task force in 2004 to examine the DC Public Library (DCPL). In November 2006 the Mayor’s Task Force Report was released, envisioning that:

“Revitalized libraries will offer fresh collections of current books and media, useful standard publications, multilingual materials, GED and SAT practice books, historic documents and records, pertinent online databases, and digital content.”

Unfortunately, listing digital content last is symbolic of the vision for DC libraries. In the 370 page Technical Report, e-books are mentioned only seven times in reference to the future of DCPL collections. Here some of the few excerpts that lay out technology vision for the DCPL in 2010 and beyond:

“[Page 21]: The library should license digital content and make it available to registered borrowers whether they are in the library or using the collection from their home. E-books, digital audiobooks, videos-on-demand, and other digital content should be available for downloading to a customer’s personal computer, PDA,or MP3 player. …

[Page 60]: A ‘virtual branch’ is fast becoming a necessary facility for successful public libraries serving large populations. … A virtual branch can be a full-service location for searching licensed electronic databases, getting answers through an interactive reference service, downloading digital books and audiovisual content, using learning software, and participating in online programs such as presentations and discussions about books and topics of current interest. Also, items in the library’s physical collections can be reserved and, when available, shipped to the user – with any fees charged to the user’s account or credit card.”

It all sounds pretty high tech until you hit on idea of shipping books around the city. In any case, if you’ve lived in DC for long, the vision of a virtual branch probably sounds like science fiction. In fact, the 2010 vision for the DCPL is based on technology that has been used in public libraries around the country for years.
.

Playing Catch Up

Several companies including NetLibrary, Overdrive, and Ebrary already have partnerships with public libraries around the world. All three offer a system for libraries to purchase e-books and popular audio book titles, though the technology is not cheap. NetLibrary has a rather shocking price structure. According to LibraryJournal.com:

“NetLibrary… now with more than 100,000 titles, has maintained its one book–one user access model. The company offers two primary purchase models based on title-by-title selection: libraries can subscribe to an ebook by paying the list price of the book, plus an annual access fee of 15 percent of the list price, or libraries can ‘own’ an ebook by paying the list price plus a one-time access fee of 55 percent of the list price.”

Unfortunately, the lack of a significant discount for subscriptions will encourage libraries to purchase e-books. If the library purchases a title, e-catalogs lose many of the advantages of digitization. For example, if econo-star Steven Levitt is heading to Seattle to talk about Freakonomics and 100 people want to read his e-book at the same time, they’ll have to wait. The King County Library System can only lend out 5 of his e-books at a time. E-books are made of 1s an 0s, not hardwoods and glue, so why not simply pay the publisher every time the virtual book is checked out?

Although the NetLibrary subscription model is extortionary, subscription e-books are the ideal model for libraries. E-books require few human resources: three weeks after you check out an e-book, it is automatically “returns” itself to the library by deleting itself (spooky, eh?). A subscription service ensures that libraries pay publishers only when e-books are checked out, so no dollars are wasted on lonely, unread books sitting on dusty shelves. Best of all, 10 of your best friends can check out a copy of the e-book at the same time, eliminating excuses in your book club.

While a subscription service might seem to favor the publishers of bestsellers, it also provides increased exposure to authors of obscure or out-of print books - the so-called long tail. If libraries pay only when e-books are loaned, then there is no reason to limit the size of their virtual catalog. A proper e-book subscription service that costs a public library no upfront fees also reduces guesswork in collections management, enabling a smooth transition in budgeting for a dedicated e-library.
.

Leapfrog: DC Should Pursue the First True Virtual Public Library System

DC Public Libraries suffer from several problems that make them an ideal test case for launching a real virtual library: huge deferred maintenance costs, an aging central library, and a population disillusioned by years of neglect to local libraries.

DCPL is too far behind the technology curve to play catch up. Instead they should leapfrog technologies. Statastic proposes that new DC Mayor Adrian Fenty and DCPL Director Ginnie Cooper consider a bold experiment in virtual collections.

The DCPL should start phasing out the acquisition of new paper books in 2008, with the goal of e-books making up no less than 90% of new acquisitions 2013. By 2017, the DCPL should have digitized 90% of its existing collection and sold the millions of hardback books in its stacks to help generate revenue. This will reduce required square footage - and overhead costs - of neighborhood libraries, eliminate the frustration of missing books, and create more space for the computer terminals that are sorely lacking.

.

Advantages to the DC Virtual Library

One of the most obvious advantages to a virtual library is the 24-hour access to books that might otherwise be checked out or unavailable. It could also expand the size of the DCPL available collection from 2.7 million volumes to as many as 32 million - every title in the WorldCat system. More titles mean more attention from residents, which increases reading and circulation.

In a city with 17% of its residents and 30% of its children living in poverty, it might seem that DC is not well-suited for e-books. After all, how would someone living in poverty afford a $300 e-reader? And how would they download a book with access to the Internet? Statastic expects the prices of e-readers featuring e-ink to drop to less than $75 within 5 years (we already have $100 laptops), and less than $40 by 2017.

Under this plan, the DCPL would phase in heavily subsidized or free e-readers for every low income DC resident. Children could also use these e-readers in the public schools where textbooks are in such short supply making it impossible to assign homework from textbooks. And assigning a hot new technology like e-readers to under-privileged citizens might just spark their interest in reading.

With the advent of Google Books, do we need a virtual public library? Many DC residents with their own e-readers and home Internet access will soon have access to millions of Google e-books. But the digital divide is real and if public libraries aren’t centrally involved in digitization of books, the gap will widen. As the high-income, early adopters turn toward e-books, wealthy taxpayers might see less value in funding the DC Public Library System (if this is even possible). This would exacerbate already grave funding shortfalls, leaving an underclass with an ignored and outmoded library system.

.

E-book Public Library Budget

Free e-readers, millions of e-books… sounds expensive, doesn’t it? Actually it could save taxpayers millions of dollars. The DCPL proposed budget for 2007 is $43 million, of which $19 million is for reference and library collection services. Facilities make up another $9.4 million. All three of these categories will have enormous savings if transitioned to an e-books catalog:

• 1. Book Sale: Bound books would be liquidated, with proceeds dedicated to digitizing rare books and subscription to an e-books catalog.
• 2. Deferred Maintenance: Neighborhood libraries that have been already been closed or that suffer from huge deferred maintenance costs would be sold and replaced by an increased number of smaller, leaner e-libraries that offer better neighborhood access to computer workstations, distribution of e-reader and training and reading programs. Neighborhoods with little Internet access would be served by DCPL e-book kiosks (more on this below).
• 3. Staff reductions: With fewer books to re-shelve and a 24 hour virtual library online, collections and maintenance staff could be reduced. As D.C. Library Renaissance Project Director Robin Diener recently commented: “We have evidence of incredible abuses — people who work for [the library] and draw a salary and rarely come to work. It’s a no-work culture.”
• 4. Private Partners: The DCPL should pursue technology partners such as Google or Yahoo which will be naturally be drawn to the cutting edge, high-profile project of digitizing the library system of the nation’s capital.
• 5. Private & Public Grants: The DCPL’s innovative virtual library experiment will also attract attention from major donors such as the Bill & Melinda Gates Foundation as well as federal grants.

By 2017 when the DC Virtual Library System is in place, there will be some new expenses (2007 dollars):

• 1. New Staff: Tech savvy staff capable of managing a virtual catalog and training patrons on the use of e-books, e-readers, and online catalogs. The new, more expensive staff hired will be offset by staff reductions due to reduced maintenance and collections services (e.g. re-shelving). Reference librarians will still be necessary, though they may take on new roles dispensing valuable advice through online forums. It’s even easy to imagine 24 hour access to an online librarian.
• 2. Subsidized E-readers: Assume that students would be issued e-readers at school. The DCPL would purchase e-readers for any adult with a household income of less than $35,000 per year. Nearly one-third, or about 184,000, DC residents would qualify for free e-readers. If new e-readers are issued every 2 years and the price of e-readers averages $40 in 2017 (it is more likely to be about $15 to $20), it would cost the city about $3.7 million annually to supply free e-readers. Other ways to help defray these costs are to require a deposit or small co-payments from those above the poverty line.
• 3. E-book Subscriptions: Circulation in all DC public libraries is about 1.1 million books. If the DCPL can reach a circulation rate similar to Seattle’s, residents would be checking out 6 million e-books per year, or about 1 book per month for every reading-age citizen. If the DCPL cut a deal with publishers to pay $2 for every book checked out from its libraries, it would cost the DCPL $12 million per year to maintain its e-book collection.
• 4. E-book Kiosks: For residents in neighborhoods with limited home Internet access, DCPL could provide e-book kiosks near public areas like schools or community centers. Pre-distributed e-readers would have RFID technology that identifies the DCPL account. The customer would simply touch the e-reader to kisok to log in. After selecting a title, the user holds the e-reader next to the kiosk for a free wireless download of the e-book. Installation and maintenance could be covered by corporate advertising on the outside of the kiosk.

The total new expenses of subsidized e-readers plus e-book subscriptions is equal to about $15.6 million. Compare this to the $19 million being spent on collections and references this year alone.

It’s important to remember that not all books, documents and historical will be digitized, so there will be still be a need for a central library, whether it is the renovated Martin Luther King, Jr. Library in downtown DC or the new one proposed by the DC Mayor. Either of the central library proposals are would cost about $275 million according to city estimates. Statastic doesn’t prefer one proposal over another, but it is imperative to rethink the needs of a central library with e-books as the heart of the collection.

.

Unknowns and Opportunities

There are two unknowns in creating the world’s first truly digital library: 1) are publishers willing to negotiate lower e-books subscription rates, and 2) will the DCPL create its own e-book collection or outsource it?

One major pre-condition for this proposal to succeed is that subscription e-book prices must be negotiated downward. Publishers and authors must be made to understand that low-priced e-book “rentals” in public libraries will increase readership. Publishers will sell higher volumes of e-books, and they can count on a reliable revenue stream.

It is not cost effective for the DCPL to digitize its collection from scratch. Google has already inserted themselves into the e-book value chain and Google is far and away the leader in the number of titles digitized. In fact, Google may be the only corporate partner for creating a virtual public library. According to Jeffrey Toobin in this week’s New Yorker, because of publishers’ lawsuits against Google, they might be the last company to digitize the world’s books:

Google’s advantage may well be cemented if the company settles its lawsuits with the publishers and authors. … [Lawrence Lessig , Professor at Stanford Law School said], “The publishers will get more than the law entitles them to, because Google needs to get this case behind it. And the settlement will create a huge barrier for any new entrants in this field.”

Google will complete digitization of the 6 million books at the University of Michigan by 2010 - the same target year for the DCPL’s modernization. The DCPL should approach Google immediately to negotiate a partnership for the DC Virtual Library.

Are DC residents feeling lucky?

.

Yesterday I wrote about the shortcomings of Prahalad’s book, The Fortune at the Bottom of the Pyramid. We left with the question of whether there was even a market to discuss. Several factors make it difficult to estimate disposable income at the bottom of the pyramid (BOP). Even if it is not a fortune, there is likely much more than a nickel a day of disposable income amongst the world’s 4 billion poor.

Most people in extreme poverty live in rural areas and derive much of their diet from subsistence farming. This means that relatively little of their income is spent of food. In family or tribe-oriented societies, there is also an income smoothing effect. Kinship networks, for example, mean that if one person in a family has a high-paying position in the government, many in the family will benefit. In addition, income such as flows from non-governmental aid, international transfers from foreign nationals living abroad, and the grey economy may be under-reported in GNP figures.

In response to Karnani’s paper, the WRI’s NextBillion.net noted that:

BOP households collectively spend money, lots of it, on a wide variety of goods and services, and are clearly willing to pay for services such as connectivity, clean water, financial services, energy, health care, and education for their children, as well as food, housing, and consumer goods. The BOP is already an economic actor, not just a passive, dependent group, and its collective actions define a market.

So there let’s assume that there is indeed a market of billions at the bottom of the pyramid. Should companies try to reach it? Karnani cautions that viewing the BOP as a vast market of micro-consumers is “potentially a dangerous delusion.” Let’s look more closely at his argument.
.

.

Distribution and Economies of Scale

Concerned about the apparent gullibility of multinational corporations (MNCs), Karnani warns that:

“Not only is the BOP market quite small, it is unlikely to be very profitable, especially for a large company. The costs of serving the markets at the bottom of the pyramid are very high…. This increases distribution and marketing costs and makes it difficult to exploit economies of scale. Weak infrastructure (transportation, communication, media, and legal) further increases cost of doing business.”

Two words: Coke and Guinness. Both have very deep penetration in West Africa. Granted these are not going to improve the health and well-being of the BOP (though Guinness bottles do read, “Guinness is Good for You“). Somehow these MNCs have overcome the challenge of distributing and marketing their products across a large geographic area.

Regardless of infrastructure and marketing costs, the market will help align buyers and sellers if the price is right for each.
.

.

Reducing Prices at the Bottom of the Pyramid

Prahalad’s thesis hinges on the idea that attracting more competition to the BOP will drive down product prices, thus freeing up their disposable income for other purchases. This is basically how Wal-Mart has made low-income Americans feel richer even as real income has stagnated over the last decade. But Karnani takes issue with Prahalad’s assertion that the private sector can deliver high quality goods to the world’s desperately poor at competitive prices:

“There are only three ways to reduce prices: 1) reduce profits, 2) reduce costs without reducing quality, and 3) reduce costs by reducing quality…. the only realistic way to reduce price is to reduce cost. The BOP proposition is adamant that we should not reduce quality in this process.

“Unless all the current producers are grossly inefficient, the only way to reduce cost… without reducing quality will always require a significant improvement in technology. Good examples of this are found in the areas of computers, telecommunications and various electronic products. It is difficult to find examples of such dramatic cost reduction in other product categories. It is not surprising that the BOP proposition repeatedly uses these same examples. We should also note that the ultimate impact on the real income of the poor due to these major price reductions is quite low because the poor spend only a small part of their income on such electronic products. The poor spend over 80% of their income on food, clothing and fuel – products that have not benefited from such dramatic technological changes in a long time.”

Let’s evaluate that last statement and have a look at how technology might help deliver improved food, clothing, fuel, and public health.

Food: There are constant improvements in pest-resistant crops, hybrid seeds, or high volume animal husbandry. Many famers in Africa still till individual family farms by hand. Certainly technology could help them improve efficiency which would lead to lower prices.

And technology improvements in computers and telecommunications do not exist in a vacuum. There are numerous positive spillover effects that affect the BOP as producers. The Washington Post recently reported that cell phones in Congo have enabled farmers and fishermen to “…use text messaging to check market prices, eliminating middlemen and increasing profits — and preventing long trips to the market on days it is canceled.” So a technology unrelated to agriculture has helped farmers saved on input prices (transport to the market on days when it’s canceled) and output prices.

Clothes: Apparel prices have tumbled over the past decade. Much of this is due to reduced quotas on Chinese apparel imports in the U.S. and Europe. Thus, the assertion that “the only way to reduce cost… without reducing quality will always require a significant improvement in technology” is inaccurate. Clothing prices have dropped as a result of trade policy, not an improvement in technology. This does have a trickle-down effect for the world’s poor.

Fuel: Fuel has indeed become more expensive. Women have to scavenge farther for firewood. Oil prices lead Nigerians into the deadly practice of siphoning off crude oil from pipelines running through their villages. But technology can improve access to energy sources. Military applications such as SkyBuilt mobile solar power could find a market at the BOP helping medical centers or providing a short term power source for harvesting and processing crops.

Public Health: As patents expire on novel drugs, cheaper generic drugs will enter these markets. Playpump is an innovative approach to water delivery. LifeStraw promises to exploit economies of scale in order to drive down prices for its personal water filtration device.

Technology: Last week, the Wall Street Journal ran a front-page story about an American entrepreneur, Greg Wyler, who was building an Internet infrastructure in Rwanda. The focus of Terracom is to first focus on market access, then profits. Mr. Wyler might disagree with Karanani’s ideas about providing a quality product at reasonable prices for the poor. He is quoted as saying, “We’re on a mission here to see what happens when we drive prices down and quality up.”

And lest you think that Rwanda is an obvious market for an outside investor, have a look at the graph at the right.

As WRI writes in response to Karnani’s critique:

The pertinent development question is whether the BOP is well served by the present (often informal) markets, and whether there are unmet needs that could be better served by more competitive markets and broader participation by the legitimate private sector.

I believe that private sector innovation help can drive prices lower, maintain or increase quality, and help deliver goods that result in better livelihoods for those at the bottom of the pyramid. But what if multinationals start marketing products that the poor don’t need? Are BOP consumers rational economic actors? Or is Karnani correct when he says that, “The problem is that the poor often make choices that are not in their own self interest.”

More on that soon…

In May, 2006 the MIT Media Lab unveiled its first working prototype of the $100 laptop. The non-profit One Laptop per Child (OLPC) was spun out of the Media Lab to manage the design. The result is small, brightly-colored and rugged laptop that may cost as much as $140. Designed to the specifications of the world’s poorest children, they are the great hope for narrowing the global digital divide.

Before they ship the final product in 2007, OLPC will pilot prototypes in the six countries listed in the chart below. The plan is for the governments of developing nations to purchase millions of these laptops and distribute them to children through their schools.

While I applaud the goal of providing an ultra-affordable laptop to the bottom of the pyramid market, I do worry about the re-sale of these high-value items. OLPC has eventual plans to create a secondary market for the sale of $100 laptops in developed world. From the OLPC FAQs:

Will OLPC spin-off a commercial subsidiary?

The idea is that a commercial subsidiary could manufacture and sell a variation of the OLPC in the developed world. These units would be marked up so that there would be a significant profit which can be plowed into providing more units in countries who cannot afford the full cost of one million machines.

The discussions around this have talked about a retail price of 3× the cost price of the units.

$100 in Nigeria is the equal to nearly two months income. To give Americans a sense of how much $100 is to the average Nigerian, imagine sending your 8 year old to school with a $6,000 laptop. Now imagine living in a country with an epidemic of corruption, in a shanty with no electricity or running water. If laptops were selling for $300 in developed nations, it would provide a strong, and unfortunate, incentive for Nigerian parents to sell their children’s laptops.

Leapfrogging technologies is a worthy goal, but OLPC has to make sure that the social institutions in their target markets can support the landing. They must concentrate as hard on issues such as corruption and cyclical poverty as they do on the design of motherboard and screen brightness. We’ve seen before with the example of SCANWATER that good technology will fail without first addressing underlying problems.
.

Sources: Statastic research, Wikipedia, IMF

Notes: PPP was not used because the calculation expresses $100 in an approximated nominal US income per capita. The statastic was calculated by dividing the nominal GDP per capita in these 7 countries by $100. That percentage was then multiplied by the nominal GDP per capita of the United States.

The term net neutrality has been the subject of half-page ads in the Washington Post, bumper stickers, even ironic t-shirts. Unfortunately, all of the advertising, partisanship, and hipsters may be obscuring the importance of a difficult debate about the evolution of the Internet. If the father and grandfather of the Internet can’t agree, what can the rest of us do?

Let’s try to understand the term net neutrality. When a consumer wants a good, service or information from the Internet, they surf an almost infinite number of web sites. Here’s the only catch: in order for a consumer to have access, they have to pay for the connection. In return consumers have equal access to all web sites.

But some broadband providers have started complaining that not content is alike. The consumer may be requesting enormous amount of data in a streaming movie, or a few 1s and 0s in an email message. So the broadband providers want to start charging the content producers for their bandwidth usage. Sure, the producers don’t have to pay these tolls, but if they don’t their competitors’ web sites might load faster.

Content providers assert that they are merely responding to consumers by sending the information they request. Broadband providers claim that with these new fees, they can build even bigger, faster networks that consumers will love.

So the producers of content are actively lobbying Congress to push for what they call “net neutrality.” This so-called neutrality is actually regulatory legislation that would prevent broadband providers from a discriminating against particular companies. It would apply common carrier rules to the Internet, ensuring that all Internet traffic is managed on equal terms, protecting startups that might not be able to compete with bigger, faster web sites.
But wait a second, it’s not just Internet giants such as Google that use an enormous amount of bandwidth, so do newcomers such as youtube.

Let’s say I’m at a startup and our business strategy is to stream movies faster than anyone else. We’ve allocated half our budget to pay Comcast cable for broadband priority. And then a pokey competitor sues. How long would it take the courts to determine what’s fair? Would bad network neutrality legislation stifle innovation more than the status quo? (Just ask Ted Stevens.) Many techies are opposed to net neutrality because they’re worried that the government just wouldn’t move fast enough if there were net neutrality violations.

But if broadband carriers start charging content providers for bandwidth, what’s to prevent them from acting more like monopolistic cable companies? Couldn’t they start packaging Internet “channels” based on which company is bidding the most? Maybe, but not for long. Current broadband providers are still under the supervisory eye of Department of Justice (anti-trust), FCC (content) and FTC (consumer protection).

So it seems to come down to the market of broadband providers. If our goal is consumer choice and innovation, then we need more competition between the carriers. How much is there? Only about 50% of Americans have two or more broadband options.

How does the U.S. compare to the rest of the world? Let’s have a look at the statastic. Even though the U.S. is twelfth in the percentage of the population with broadband connectivity, we do have three competing broadband options: cable, DSL, and satellite. As a nation, we have a healthy broadband competition relative to the rest of the world… even if it doesn’t seem that way in your neck of the woods.

Sources: Statastic research, OECD

The Internet has been available in the developing world almost as long as it’s been here in the U.S. Internet cafes were popping up in Cameroon in the mid 1990s before the local Peace Corps volunteers even knew how to use them. Penetration rates, however, lag predictably behind the richer countries in the north. But the lack of telecommunications infrastructure is something of a blessing in disguise: developing nations have the potential to leapfrog technologies. Cell phones and VOIP prove easier than installing costly land lines, and there’s no need for telephone poles and copper cable if governments can create WiFi and WiMAX zones around burgeoning urban areas.

Wired Magazine recently featured a map with average prices for one hour of online access in Internet cafes around the world. Statastic used the average hourly price as a percentage of daily wages to provide a glimpse into the state of Internet access in a selection of low to middle income countries.

The chart below begs several questions. Could lowering the cost of public Internet access lead to higher usage rates? What is the demographic profile of the average Internet user in the developing world? Should multi-lateral donors subsidize the cost of public Internet access?

Among this small sample, D.R. Congo, Nigeria and Kenya are the three most expensive places for locals to access the Internet, relative to income. They also have some of the lowest usage rates. But these countries have several other characteristics in common: low literacy, high rates of corruption, and a high level of inequality. These countries may simply have a limited number of Internet cafes that cater to tourists, corrupt officials and the wealthy locals who are lucky enough to have an education and a job.

Brazil’s usage rates are surprisingly high. Perhaps Brazil’s high inequality can help explain how 14% of Brazilians have regular access to the Internet despite the fact that one hour in an Internet café costs nearly one sixth of average daily wages. Just who are those fortunate 14%?

Sources:
http://internetworldstats.com
WIRED Magazine, May 2006