The Alchian-Allen Theorem and the Economics of Internet Animals

Abstract

Economics of Cute          There are many ways to study cute: for example, neuro-biology (cute as adaptation); anthropology (cute in culture); political economy (cute industries, how cute exploits consumers); cultural studies (social construction of cute); media theory and politics (representation and identity of cute), and so on. What about economics? At first sight, this might point to a money-capitalism nexus (“the cute economy”), but I want to argue here that the economics of cute actually works through choice interacting with fixed costs and what economists call ”the substitution effect”. Cute, in conjunction with the Internet, affects the trade-offs involved in choices people make. Let me put that more starkly: cute shapes the economy. This can be illustrated with internet animals, which at the time of writing means Grumpy Cat.   I want to explain how that mechanism works – but to do so I will need some abstraction. This is not difficult – a simple application of a well-known economics model, namely the Allen-Alchian theorem, or the  “third law of demand”. But I am going to take some liberties in order to represent that model clearly in this short paper. Specifically, I will model just two extremes of quality (“opera” and  “cat videos”) to represent end-points of a spectrum. I will also assume that the entire effect of the internet is to lower the cost of cat videos. Now obviously these are just simplifying assumptions “for the purpose of the model”. And the purpose of the model is to illuminate a further aspect of how we might understand cute, by using an economic model of choice and its consequences. This is a standard technique in economics, but not so in cultural studies, so I will endeavour to explain these moments as we go, so as to avoid any confusion about analytic intent. The purpose of this paper is to suggest a way that a simple economic model might be applied to augment the cultural study of cute by seeking to unpack its economic aspect. This can be elucidated by considering the rise of internet animals as a media-cultural force, as epitomized by “cat videos”. We can explain this through an application of price theory and the theory of demand that was first proposed by Armen Alchian and William Allen. They showed how an equal fixed cost that was imposed to both high-quality and low-quality goods alike caused a shift in consumption toward the higher-quality good, because it is now relatively cheaper. Alchian and Allen had in mind something like transport costs on agricultural goods (such as apples). But it is also true that the same effect works in reverse (Cowen), and the purpose of this paper is to develop that logic to contribute to explaining how certain structural shifts in production and consumption in digital media, particularly the rise of blog formats such as Tumblr, a primary supplier of kittens on the Internet, can be in part understood as a consequence of this economic mechanism. There are three key assumptions to build this argument. The first is that the cost of the internet is independent of what it carries. This is certainly true at the level of machine code, and largely true at higher levels. What might be judged aesthetically high quality or low quality content – say of a Bach cantata or a funny cat video – are treated the same way if they both have the same file size. This is a physical and computational aspect of net-neutrality. The internet – or digitization – functions as a fixed cost imposed regardless of what cultural quality is moving across it. Second, while there are costs to using the internet (for example, in hardware or concerning digital literacy) these costs are lower than previous analog forms of information and cultural production and dissemination. This is not an empirical claim, but a logical one (revealed preference): if it were not so, people would not have chosen it. The first two points – net neutrality and lowered cost – I want to take as working assumptions, although they can obviously be debated. But that is not the purpose of the paper, which is instead the third point – the  “Alchian-Allen theorem”, or the third fundamental law of demand. The Alchian-Allen Theorem  The Alchian-Allen theorem is an extension of the law of demand (Razzolini et al) to consider how the distribution of high quality and low quality substitutes of the same good (such as apples) is affected by the imposition of a fixed cost (such as transportation). It is also known as the  “shipping the good apples out” theorem, after Borcherding and Silberberg explained why places that produce a lot of apples – such as Seattle in the US – often also have low supplies of high quality apples compared to places that do not produce apples, such as New York. The puzzle of  “why can’t you get good apples in Seattle?” is a simple but clever application of price theory. When a place produces high quality and low quality items, it will be rational for those in faraway places to consume the high quality items, and it will be rational for the producers to ship them, leaving only the low quality items locally.Why? Assume preferences and incomes are the same everywhere and that transport cost is the same regardless of whether the item shipped is high or low quality. Both high quality and low quality apples are more expensive in New York compared to Seattle, but because the fixed transport cost applies to both the high quality apples are relatively less expensive. Rational consumers in New York will consume more high quality apples. This makes fewer available in Seattle.Figure 1: Change in consumption ratio after the imposition of a fixed cost to all apples Another example: Australians drink higher quality Californian wine than Californians, and vice versa, because it is only worth shipping the high quality wine out. A counter-argument is that learning effects dominate: with high quality local product, local consumers learn to appreciate quality, and have different preferences (Cowen and Tabarrok).The Alchian-Allen theorem applies to any fixed cost that applies generally. For example, consider illegal drugs (such as alcohol during the US prohibition, or marijuana or cocaine presently) and the implication of a fixed penalty – such as a fine, or prison sentence, which is like a cost – applied to trafficking or consumption. Alchian-Allen predicts a shift toward higher quality (or stronger) drugs, because with a fixed penalty and probability of getting caught, the relatively stronger substance is now relatively cheaper. Empirical work finds that this effect did occur during alcohol prohibition, and is currently occurring in narcotics (Thornton Economics of Prohibition, "Potency of illegal drugs").Another application proposed by Steven Cuellar uses Alchian-Allen to explain a well-known statistical phenomenon why women taking the contraceptive pill on average prefer  “more masculine” men. This is once again a shift toward quality predicted on falling relative price based on a common ‘fixed price’ (taking the pill) of sexual activity. Jean Eid et al show that the result also applies to racehorses (the good horses get shipped out), and Staten and Umbeck show it applies to students – the good students go to faraway universities, and the good student in those places do the same. So that’s apples, drugs, sex and racehorses. What about the Internet and kittens?Allen-Alchian Explains Why the Internet Is Made of CatsIn analog days, before digitization and Internet, the transactions costs involved with various consumption items, whether commodities or media, meant that the Alchian-Allen effect pushed in the direction of higher quality, bundled product. Any additional fixed costs, such as higher transport costs, or taxes or duties, or transactions costs associated with search and coordination and payment, i.e. costs that affected all substitutes in the same way, would tend to make the higher quality item relatively less expensive, increasing its consumption.But digitisation and the Internet reverse the direction of these transactions costs. Rather than adding a fixed cost, such as transport costs, the various aspects of the digital revolution are equivalent to a fall in fixed costs, particularly access.These factors are not just one thing, but a suite of changes that add up to lowered transaction costs in the production, distribution and consumption of media, culture and games. These include: The internet and world-wide-web, and its unencumbered operation The growth and increasing efficacy of search technology Growth of universal broadband for fast, wide band-width access Growth of mobile access (through smartphones and other appliances) Growth of social media networks (Facebook, Twitter; Metcalfe’s law) Growth of developer and distribution platforms (iPhone, android, iTunes) Globally falling hardware and network access costs (Moore’s law) Growth of e-commerce (Ebay, Amazon, Etsy) and e-payments (paypal, bitcoin) Expansions of digital literacy and competence Creative commons These effects do not simply shift us down a demand curve for each given consumption item. This effect alone simply predicts that we consume more. But the Alchian-Allen effect makes a different prediction, namely that we consume not just more, but also different.These effects function to reduce the overall fixed costs or transactions costs associated with any consumption, sharing, or production of media, culture or games over the internet (or in digital form). With this overall fixed cost component now reduced, it represents a relatively larger decline in cost at the lower-quality, more bite-sized or unbundled end of the media goods spectrum. As such, this predicts a change in the composition of the overall consumption basket to reflect the changed relative prices that these above effects give rise to. See Figure 2 below (based on a blog post by James Oswald). The key to the economics of cute, in consequence of digitisation, is to follow through the qualitative change that, because of the Alchian-Allen effect, moves away from the high-quality, highly-bundled, high-value end of the media goods spectrum. The  “pattern prediction” here is toward more, different, and lower quality: toward five minutes of  “Internet animals”, rather than a full day at the zoo. Figure 2: Reducing transaction costs lowers the relative price of cat videos Consider five dimensions in which this more and different tendency plays out. Consumption  These effects make digital and Internet-based consumption cheaper, shifting us down a demand curve, so we consume more. That’s the first law of demand in action: i.e. demand curves slope downwards. But a further effect – brilliantly set out in Cowen – is that we also consume lower-quality media. This is not a value judgment. These lower-quality media may well have much higher aesthetic value. They may be funnier, or more tragic and sublime; or faster, or not. This is not about absolute value; only about relative value. Digitization operating through Allen-Alchian skews consumption toward the lower quality ends in some dimensions: whether this is time, as in shorter – or cost, as in cheaper – or size, as in smaller – or transmission quality, as in gifs. This can also be seen as a form of unbundling, of dropping of dimensions that are not valued to create a simplified product.So we consume different, with higher variance. We sample more than we used to. This means that we explore a larger information world. Consumption is bite-sized and assorted. This tendency is evident in the rise of apps and in the proliferation of media forms and devices and the value of interoperability.ProductionAs consumption shifts (lower quality, greater variety), so must production. The production process has two phases: (1) figuring out what to do, or development; and (2) doing it, or making. The world of trade and globalization describes the latter part: namely efficient production. The main challenge is the world of innovation: the entrepreneurial and experimental world of figuring out what to do, and how. It is this second world that is radically transformed by implications of lowered transaction costs.One implication is growth of user-communities based around collaborative media projects (such as open source software) and community-based platforms or common pool resources for sharing knowledge, such as the  “Maker movement” (Anderson 2012). This phenomenon of user-co-creation, or produsers, has been widely recognized as an important new phenomenon in the innovation and production process, particularly those processes associated with new digital technologies. There are numerous explanations for this, particularly around preferences for cooperation, community-building, social learning and reputational capital, and entrepreneurial expectations (Quiggin and Potts, Banks and Potts). Business Models  The Alchian-Allen effect on consumption and production follows through to business models. A business model is a way of extracting value that represents some strategic equilibrium between market forms, organizational structures, technological possibilities and institutional framework and environmental conditions that manifests in entrepreneurial patterns of business strategy and particular patterns of investment and organization. The discovery of effective business models is a key process of market capitalist development and competition. The Alchian-Allen effect impacts on the space of effective viable business models. Business models that used to work will work less well, or not at all. And new business models will be required. It is a significant challenge to develop these  “economic technologies”. Perhaps no less so than development of the physical technologies, new business models are produced through experimental trial and error. They cannot be known in advance or planned. But business models will change, which will affect not only the constellation of existing companies and the value propositions that underlie them, but also the broader specializations based on these in terms of skill sets held and developed by people, locations of businesses and people, and so on. New business models will emerge from a process of Schumpeterian creative destruction as it unfolds (Beinhocker).    The large production, high development cost, proprietary intellectual property and systems based business model is not likely to survive, other than as niche areas. More experimental, discovery-focused, fast-development-then-scale-up based business models are more likely to fit the new ecology. Social Network Markets & Novelty Bundling MarketsThe growth of variety and diversity of choice that comes with this change in the way media is consumed to reflect a reallocation of consumption toward smaller more bite-sized, lower valued chunks (the Alchian-Allen effect) presents consumers with a problem, namely that they have to make more choices over novelty. Choice over novelty is difficult for consumers because it is experimental and potentially costly due to risk of mistakes (Earl), but it also presents entrepreneurs with an opportunity to seek to help solve that problem. The problem is a simple consequence of bounded rationality and time scarcity. It is equivalent to saying that the cost of choice rises monotonically with the number of choices, and that because there is no way to make a complete rational choice, agents will use decision or choice heuristics. These heuristics can be developed independently by the agents themselves through experience, or they can be copied or adopted from others (Earl and Potts). What Potts et al call  “social network markets” and what Potts calls  “novelty bundling markets” are both instances of the latter process of copying and adoption of decision rules. Social network markets occur when agents use a  “copy the most common” or  “copy the highest rank” meta-level decision rule (Bentley et al) to deal with uncertainty. Social network markets can be efficient aggregators of distributed information, but they can also be path-dependent, and usually lead to winner-take all situations and dynamics. These can result in huge pay-offs differentials between first and second or fifth place, even when the initial quality differentials are slight or random. Diversity, rapid experimentation, and  “fast-failure” are likely to be effective strategies. It also points to the role of trust and reputation in using adopted decision rules and the information economics that underlies that: namely that specialization and trade applies to the production and consumption of information as well as commodities. Novelty bundling markets are an entrepreneurial response to this problem, and observable in a range of new media and creative industries contexts. These include arts, music or food festivals or fairs where entertainment and sociality is combined with low opportunity cost situations in which to try bundles of novelty and connect with experts. These are by agents who developed expert preferences through investment and experience in consumption of the particular segment or domain. They are expert consumers and are selling their  “decision rules” and not just the product. The more production and consumption of media and digital information goods and services experiences the Alchian-Allen effect, the greater the importance of novelty bundling markets.   Intellectual Property & Regulation A further implication is that rent-seeking solutions may also emerge. This can be seen in two dimensions; pursuit of intellectual property (Boldrin and Levine); and demand for regulations (Stigler). The Alchian-Allen induced shift will affect markets and business models (and firms), and because this will induce strategic defensive and aggressive responses from different organizations. Some organizations will seek to fight and adapt to this new world through innovative competition. Other firms will fight through political connections.     Most incumbent firms will have substantial investments in IP or in the business model it supports. Yet the intellectual property model is optimized for high-quality large volume centralized production and global sales of undifferentiated product. Much industrial and labour regulation is built on that model. How governments support such industries is predicated on the stability of this model. The Alchian-Allen effect threatens to upset that model. Political pushback will invariably take the form of opposing most new business models and the new entrants they carry. Conclusion I have presented here a lesser-known but important theorem in applied microeconomics – the Alchian-Allen effect – and explain why its inverse is central to understanding the evolution of new media industries, and also why cute animals proliferate on the Internet. The theorem states that when a fixed cost is added to substitute goods, consumers will shift to the higher quality item (now relatively less expensive). The theorem also holds in reverse, when a fixed cost is removed from substitute items we expect a shift to lower quality consumption. The Internet has dramatically lowered fixed costs of access to media consumption, and various development platforms have similarly lowered the costs of production. Alchian-Allen predicts a shift to lower-quality, ”bittier” cuter consumption (Cowen). References Alchian, Arman, and William Allen. Exchange and Production. 2nd ed. Belmont, CA: Wadsworth, 1967. Anderson, Chris. Makers. New York: Crown Business, 2012. Banks, John, and Jason Potts. "Consumer Co-Creation in Online Games." New Media and Society 12.2 (2010): 253-70. Beinhocker, Eric. Origin of Wealth. Cambridge, Mass.: Harvard University Press, 2005. Bentley, R., et al. "Regular Rates of Popular Culture Change Reflect Random Copying." Evolution and Human Behavior 28 (2007): 151-158. Borcherding, Thomas, and Eugene Silberberg. "Shipping the Good Apples Out: The Alchian and Allen Theorem Reconsidered." Journal of Political Economy 86.1 (1978): 131-6. Cowen, Tyler. Create Your Own Economy. New York: Dutton, 2009. (Also published as The Age of the Infovore: Succeeding in the Information Economy. Penguin, 2010.) Cowen, Tyler, and Alexander Tabarrok. "Good Grapes and Bad Lobsters: The Alchian and Allen Theorem Revisited." Journal of Economic Inquiry 33.2 (1995): 253-6. Cuellar, Steven. "Sex, Drugs and the Alchian-Allen Theorem." Unpublished paper, 2005. 29 Apr. 2014 ‹http://www.sonoma.edu/users/c/cuellar/research/Sex-Drugs.pdf›.Earl, Peter. The Economic Imagination. Cheltenham: Harvester Wheatsheaf, 1986. Earl, Peter, and Jason Potts. "The Market for Preferences." Cambridge Journal of Economics 28 (2004): 619–33. Eid, Jean, Travis Ng, and Terence Tai-Leung Chong. "Shipping the Good Horses Out." Wworking paper, 2012. http://homes.chass.utoronto.ca/~ngkaho/Research/shippinghorses.pdf Potts, Jason, et al. "Social Network Markets: A New Definition of Creative Industries." Journal of Cultural Economics 32.3 (2008): 166-185. Quiggin, John, and Jason Potts. "Economics of Non-Market Innovation & Digital Literacy." Media International Australia 128 (2008): 144-50. Razzolini, Laura, William Shughart, and Robert Tollison. "On the Third Law of Demand." Economic Inquiry 41.2 (2003): 292–298. Staten, Michael, and John Umbeck. “Shipping the Good Students Out: The Effect of a Fixed Charge on Student Enrollments.” Journal of Economic Education 20.2 (1989): 165-171. Stigler, George. "The Theory of Economic Regulation." Bell Journal of Economics 2.1 (1971): 3-22. Thornton, Mark. The Economics of Prohibition. Salt Lake City: University of Utah Press, 1991.Thornton, Mark. "The Potency of Illegal Drugs." Journal of Drug Issues 28.3 (1998): 525-40.