The world’s fastest supercomputer is widely acclaimed to be 天河-2 (pronounced “tianhe-2” and meaning, literally, “Heaven River-2,” or “Milky Way-2.”) Developed by 1,300 scientists and engineers at a cost of 2.4 billion yuan (390 million US dollars), and deployed for the purpose of national defence in the National Supercomputer Centre of Guangzhou, China, this 80,000-processor behemoth boasts a performance of almost 34 PFLOPS (or, 34 x 10^15 floating operations per second.)
It is laughable to think that the lowly personal computer (PC) would have any chance of competing against such a titan. However, if distributed computing systems are included in the comparison, the fastest supercomputer in the world is actually Folding@home (FAH, or F@h) a shared pool of approximately 154,000 processors whose combined performance is 36 PFLOPS – a staggering two thousand million million floating operations per second faster than 天河-2. To achieve this incredible speed, the system puts to work the unused processing capacity of PCs owned by tens of thousands of loosely-affiliated volunteers who have installed F@h software on their computers for no commercial gain. Operated by Stanford University, F@h is dedicated to elucidating the thorny problem of protein folding, misfolding, and related diseases.
The potential advantage of shared over single-owner resources is not limited, as illustrated above, to competition between pooled, privately owned capacity and state-managed solutions, but also poses a meaningful threat to the corporation. For example, AirBnB, which makes spare bedrooms the world over available for rent, claims an inventory of approximately 1 million rooms. By comparison, the largest hotel groups in the world – InterContinental, Hilton, and Marriott – each have approximately 700,000 available rooms.
According to Yochai Benkler of Yale Law School, shareable goods are those that are (1) technically “lumpy” and (2) of mid-grained “granularity.” By “lumpy,” he means indivisible, or quantized. For example, it is not possible to buy computer processing power in less than a discrete unit, even if not all of that computational power is required by the buyer. The same goes for automobiles. One cannot own less than a whole car, even if that car sits idle in a garage 22 hours out of every day in the year. Benkler elaborates on granularity thus:
“Lumpy goods can, in turn, be fine, mid-, or large-grained. A large-grained good is one that is so expensive that it can only be used by aggregating demand for it. Industrial capital equipment, like steam engines, is of this type. Fine-grained goods are of a granularity that allows consumers to buy precisely as much of the goods as has the amount of capacity they require, such as a donut, or a cup of coffee. Mid-grained goods are small enough for an individual to justify buying for her own use.”
Examples of mid-grained goods in the developed world are PCs and automobiles. Benkler proposes that goods of this granularity are (i) widely owned by private individuals, and (ii) exhibit systematic excess capacity with respect to their owners’ needs.
The problem of re-allocating this slack efficiently to non-owners can be solved by one of two routes: secondary markets, or social sharing (if state-based management of private resources is explicitly excluded). Benkler proposes that the owner of a mid-grained resource with excess capacity (an “agent”) must go through a series of decisions like those outlined in Figure 1 in order to determine whether or not to exclude other people from this resource, and to what degree, based upon consideration of both the transaction and opportunity costs associated with each fork in the process.
Figure 1. Decision tree by which an agent may determine whether and how to reallocate unused capacity in a mid-grained resource
In Benkler’s model of social sharing, the owner of a mid-grained resource with excess capacity derives no monetary benefit from sharing it with others. (She may, however, receive a social benefit.) By contrast, what has become known as the “sharing economy” explicitly includes the possibility for owners to profit from their underused resources. In other words, the sharing economy conflates the secondary market with social sharing.
In the sharing economy, individuals, corporations, non-profit organizations, and governments are empowered by free-flowing information to reallocate excess capacity in goods or services by resale, barter, lending, gifting, and sharing. The transaction costs incurred by agents in the partial exclusion of others from their slack resources have been driven down dramatically over the last several years by the rise of internet-based services that connect buyers and sellers, which has resulted in a thriving ecosystem at the bottom of Benkler’s decision tree. Because the reach of products and services in the sharing economy extend beyond singular ownership, participants have been described as participating in “disownership” and “collaborative consumption”.
Forbes Magazine estimated that “revenue flowing through the sharing economy directly into people’s wallets” exceeded 3.5 billion US dollars in 2013. It predicted annual growth of 25 % and foresaw the rise of a “disruptive economic force.” Investors appear to agree and have placed bets in the hundreds of millions of dollars on start-up companies like Uber and AirBnB. It is likely that the sharing economy will have a major societal impact.
The rise of the sharing economy has been driven by:
- The Internet and social media, which dramatically reduce the transaction costs associated with establishing connections and developing trust between buyers and sellers (through bi-directional rating systems). Online marketplaces also allow secure, frictionless payments.
- Sluggishness in the global economy since 2008 and rising unemployment in the developed world, which have combined to make it necessary for many people (i) to seek additional sources of income by monetizing their underused goods; and (ii) to reduce costs by buying used goods and sharing services.
- Enhanced social awareness and a desire to do good, both to the environment (by reusing physical goods and therefore reducing waste) and to people (by helping and sharing.)
Hamari et al. define collaborative consumption as “peer-to-peer-based activity of obtaining, giving, or sharing access to goods and services, coordinated through community-based online services.” In a study of 254 such services, these researchers uncovered the following kinds of transactions: sharing, new purchase, second hand purchase, renting, donating, swapping and lending/borrowing. These were broadly classified into two main categories of exchange: access over ownership and transfer of ownership, with the former predominating.