Shoegazing and the Cloud

I have been trying, off and on since yesterday, to download the new My Bloody Valentine album (mbv) but their card payment interface appears to be banjaxed and I refuse to sign up to PayPal. You can only download it from the band's own website, which is one in the eye for tax-dodgers like Amazon but a bit pointless if it doesn't actual work. I've even emailed their support address, but given they've taken 22 years to deliver this follow-up to Loveless, I'm not expecting a swift reply. Fortunately, they've also stuck the lot up on YouTube. I highly recommend "only tomorrow", "new you" and "in another way".

The experience has not just reminded me that the Internet doesn't always work, dammit, but also that we (or at least I) remain wedded to the joys of owning stuff. Even though I can quite happily stream it online I want the download and am seriously considering buying the vinyl when it appears in a few weeks (it's an analogue recording, so this is not just vinyl collector's completitis). This reminded me of Steve Wozniak's rant a few months ago about the dangers of cloud computing. Wozniak, if you didn't know, was the original engineering brains behind Apple who was marginalised because he wasn't as pretty or as psychopathic as his partner, Steve Jobs (think Chewbacca to Han Solo, only shorter). According to The Woz, "With the cloud, you don't own anything. You already signed it away. I want to feel that I own things".

With the recent demise of HMV, it's hard to argue with the proposition implicit in Wozniak's plea: in the future we will have access to much more, but we will own much less. For most people in the developed world, the twentieth century can be characterised in terms of increasing property ownership. I don't just mean in the narrow sense of owning a house, but in the wider sense of acquiring more and more goods and chattels, from furniture to toys, from cars to CDs. While the rich obviously maintained a monopoly on certain types of property, the material prosperity of capitalism, the sheer profusion of commodities, meant that our lives were increasingly organised around the making, buying and using of stuff.

Since the introduction of the iPod in 2001, which coincided with the peak in UK property ownership, this process has taken a new direction. It's not in-yer-face obvious, as we are surrounded by a lifetime's collection of books, videos, DVDs and CDs, but the future is clearly the "weightlessness" of content on demand. Even the historic growth of user devices is misleading, both because commodity deflation means they account for proportionately less of our expenditure and because there is a limit to the number of dedicated devices we can practically use (we probably peaked with e-readers). Though Apple are obviously keen on high-margin devices, and pitch them as status symbols, the general trend is towards commoditisation and disposability. The commercial model of mobile phones, and the quickening pace of obsolescence, means that we're increasingly open to the idea that we acquire our devices as part of a contract for service access and content. We're renting more, and this applies to cultural products as much as homes. Just as the growth of buy-to-let means that property ownership is concentrating, so cloud computing is leading to a concentration of massive providers such as Amazon, Google and Apple.

There are two paradoxes here. The first is that openness and choice may be leading to homogenisation and oligopoly. The second is that the demands of cloud interoperability and the environmental costs of data centres may lead to the minimisation of data and the maximisation of access - e.g. a single copy of an MP3 file (allowing for backups and mirrors) that is accessible to everyone. This is the optimal rent model: the greatest revenue from the smallest asset.

The "move to the cloud" is central to business too. For some years now, the concept of IT as a utility has been pushed both as a desirable goal ("you don't have to think about it") and a practical service delivery model ("just plug in"). In fact, cloud computing bears little resemblance to a utility. Despite the contrived appearance of multiple providers in a free market, most public utilities are physical monopolies at the sharp end: one electricity cable, one water mains, one sewer etc. The characteristics of a utility are those of a natural monopoly. Unlike utilities, IT cloud services are neither consistent nor easily substituted between providers. Salesforce.com and Google Apps are not interchangeable. The utility trope is primarily a sales pitch, but it is also subtly ideological in that it hints at semi-official status (every accounts payable function knows you pay the utility bills first). There is also the semi-conscious derogation of "hard" IT, i.e. the equating of infrastructure roles with utility workers, which leads to the perception that cloud computing facilitates a shift from blue to white-collar jobs in IT (there is little expectation of net job reductions and some belief that the aggregate impact is an increase in roles).

The real commonality between a utility and the cloud is that the service cost includes an element of rent (in the economist's sense - i.e. a royalty). Despite the loud claims made by utility companies about their investment in infrastructure and labour (water pipes and engineers), the gross profit of a utility such as water is largely rent (the exploitation of a natural resource) rather than the surplus value of labour. Though the cloud may provide access to differentiated applications, which are ultimately the product of labour, in practice a large part of the fee you are paying is rent (hence pricing tends towards per-user rather than per-transaction models). The very opaqueness and indeterminacy of the cloud (notice it's "the" not "a") makes this difficult to see, though you get a hint of the role that rent plays in modern capitalism in the cross-charging of IP royalties by multinationals, such as Amazon.

Cloud computing consolidates capital with the cloud service providers. This not only gives rise to economies of scale for them, it also means that the businesses that use the cloud are increasingly capital light. This is good insofar as it lowers the cost of business entry and avoids the need for periodic loans or share issues to raise capital, but it also reduces the scope for internal productivity gains (e.g. by upgrading equipment) and makes process innovation dependent on onerous supplier management (this is a feature of outsourcing generally and not just cloud computing). In effect, we're seeing the centralisation of capital in fewer hands. This does not mean that cloud-using businesses are less profitable, but that profits are more likely to be recycled as bonuses and share buybacks rather than capital investment. This is the nature of the "weightless economy". The weight doesn't disappear, it simply moves to one end of the corporate spectrum.

It's early days, but the trend this century appears to be towards an economy in which the elite owns an even more disproportionate share of property than today while the majority of the population pays rent to make use of it. We will have access to more stuff, so we will feel materially richer than we did in the twentieth century, but we will own less of it. One by-product of this shift from property to rent will be the need to continue earning longer or face greater relative poverty in old age, when you typically seek to rely more on assets built up over your working life. The extending of the retirement age may prove to be about more than just a shortfall in pension contributions.