From Bearer Bonds to the Blockchain: Artistic Perspectives on Digital Money

Bitcoin and the blockchain have generated an enormous amount of press, as well as investment by governments, banks and technology companies. Ruth Catlow of Furtherfield  and Ben Vickers of Serpentine Galleries, with the generous support of the Austrian Cultural Forum London organised a fantastic workshop attended by artists, representatives of public and commercial arts organisations, and technologists. The agenda was to consider what the blockchain, a poorly-understood yet politically-charged technology that means many things to many people, might mean for art and artists, and society-at-large.

Motivation

Although our audience had a specifically artistic interest, the discussion around blockchain intertwines economics, technology, and public policy, so I thought it might help to take a step back and start by thinking about money pre-bitcoin. Specifically, I wanted to un-entangle some of the philosophical and historical questions around money; not least because the nature of money itself, and long-standing assumptions in Western economies, are today being subverted by the ongoing after-shocks of the 2008 financial crisis, to wit: negative interest rates, the potential phasing out of cash, and financial repression.

Money as Social Construct

I see money as a social construct, or alternatively, a de-centralised social contract with few explicitly articulated constraints on the individual, but plenty of implicit conventions-in-use. The essence of this contract is that money gives its holder a claim on assets or labour. It generally implies at least two parties in any given transaction; as well as a belief system that most parties in the given society mostly agree upon. These ways we use money are diverse, but are closely related to metrics for the utility of the money in question: authentication, anonymity, portability, convenience, legal certainty and fungibility.

Some examples may help.

Cash

Cash appears to be the simplest form of money. U.S. Dollars represent a claim on the American Treasury via the Federal Reserve Banking system. Since dollars are no longer backed by gold reserves – in other words, if one goes to the Fed or Treasury with a briefcase of dollars, one is not generally entitled to an equivalent amount of gold coins in return – this claim on the U.S. is termed fiat money. That is, most people, domestically and globally, agree that the Fed, and therefore the Treasury, is good for its dollars, and that everyone else in the system will provide goods and services against dollars. Furthermore, most people agree that the Fed will run the U.S. economy in a reasonably responsible way – though this consensus is far from unanimous. When, and if, this confidence erodes substantially, one may see devaluation, or inflation, depending on how the relative price of dollars to foreign currency, or to domestic goods/services, respectively, have changed. A recent example on how belief systems regarding money can change: during the summer 2015 EU crisis when Greece looked likely to be ejected from the Eurozone, there was much serious talk on how a parallel currency could co-exist with the Euro (the so-called ‘New Drachma’).

Cash is more or less anonymous in use, often is free of record/receipt, and may be irreversible (i.e. no refunds). The central bank doesn’t generally keep a ledger of individuals or businesses that hold modest amounts of cash; they do keep a ledger for commercial banks, which have a special relationship with the central bank. The central bank also closely watches how much cash is in circulation on an aggregate basis (so-called M0 or base money) and monetary policy is the science, and art, of managing M0 (and its friends M1-M4), in order to meet various legislative or statutory goals: often, price stability and low unemployment. The anonymity of cash is one reason it is interesting for tax evaders and tax authorities alike. As an aside, for all the noise about offshore accounts in the Panama Papers, HMRC estimates £16.5BN of lost UK tax revenue actually comes from VAT evasion – i.e. paying one’s plumber in cash, an entirely British affair that has nothing to do with offshore centres.

The authenticity of cash is verifiable in a relatively de-centralised way, in a reasonably short period of time: an experienced eye can spot most counterfeit bills, and marker pens or UV lights can easily be used anywhere.  Cash is portable, yet is purely symbolic – for the most part, the social contract that underpins cash, isn’t spelled out in other than the most terse terms on a bill. Most of the enabling legislation, case law, and so forth, exists outside the cash note. These points will become important later in the blockchain section.

It’s worth noting that many of the features above, which fall under the rubric convenience, mean cash can also be stolen easily and thus is expensive to store, transport, and insure.

Precious Metal

I will intentionally not say much about gold and silver, as the history and cultural aspects, while fascinating, would hopelessly complicate the discussion. Indeed, most respectable commentators hold that gold is a ‘barbarous relic’ (as John Maynard Keynes’ put it in 1923) and should be consigned to the bejewelled dustbin of history. Suffice it to say: gold is anonymous, reasonably portable (relative to value), easily authenticated, requires no ledger, but like cash, is expensive to protect. As above, gold relies on an obscure social contract and judging by the price action, I’m not sure the consensus on gold’s value hasn’t changed. Gold also may suffer from the nasty habit of being declared illegal just about when it might prove useful, due to inflation or war – for instance see Executive Order 6102, signed in 1933, that criminalised private ownership of monetary gold during the Great Depression.

Bank Deposits

In contrast to the above, money sitting in a bank has no material form, until it is pulled out of an ATM. It sits in the bank’s ledger, and definitely belongs to someone – namely, the depositor. Thus it is not anonymous, and is, in the absence of fraud or error, authenticated. Nor is it portable, though one can transfer deposits to another bank.

Bank (time) deposits do, however, earn interest , while cash obviously doesn’t. More accurately, they used to earn interest – in what some will call financial repression, and others, unconventional monetary policy, retail deposits in most Western nations, and Japan, earn near-zero interest. In an increasing number of situations, certain deposits now earn negative interest rates – one pays the bank to place a deposit.

Bank deposits also exhibit double counting. In its first guise, double counting is more commonly known as fractional reserve banking. Basically, very little cash money actually exists in the banking system. A great deal of the ‘money’ in circulation is in the form of on-demand deposits at various commercial banks (so-called M1). But those deposits rarely become cash. Bluntly, when a mortgage bank disburses a loan to a home-buyer, it almost never pays out hundreds of thousands of pounds in cash; rather, there are a chain of debits and credits, respectively, of the buyer and seller deposit accounts, that result in the seller’s account going up by the price of the house. The bottom line, for the financial system as a whole, is that a tiny sliver of M0 cash supports a large volume of M1 deposits. A bank run is when all depositors want their cash out of the bank at the same time, and there isn’t enough to go around.

A more relevant example of double-counting is when a transaction actually happens, there is often a period of 1-3 days, when balances are in transition between the seller’s deposit account and that of the buyer. Again, there are a chain of transactions, often involving two or more banks (i.e. the seller’s and buyer’s), as well as the central bank, during which time deposits may well be double-counted in the accounts of multiple financial institutions. This is mostly fine, because the central bank, in theory, supervises the system, provides plenty of extra cash liquidity and, in any case, the double-counting doesn’t last for very long; however, when there is a financial crisis, this could become a problem. This double-counting problem applies to blockchain precisely because the latter is basically a simultaneous system – transactions are confirmed and settled almost instantaneously.  I would point the reader to Izabella Kaminska’s writings over at the Financial Times for more on this topic, and in general for excellent critical commentary on how various quasi-utopian aspirations of digital money square up to the sometimes brutal realities of the financial world.

Bonds and Shares

The last animals in this taxonomy are bonds and shares, of which I’ll only treat the former. Bonds combine many of the features above, namely they are a promise (i.e. debt) by a party (the issuer) to pay another party (the bondholder) in a certain currency (dollars, pounds, etc.) at a certain rate of interest (known as the coupon) on a certain date (the maturity date of the bond).

The first difference we note is that bonds have a maturity date – unlike on-demand deposits, the bondholder generally can’t get his/her money back whenever he/she wants. So they look more like time-deposits in a bank (which incidentally go into M2).

Bonds, interestingly, used to come in two flavours: bearer and registered. Registered bonds usually have no physical form: rather they are an entry on the ledger of a central record-keeping agency, a so-called custodian. There is usually a specific method of transferring them, through a so-called clearing system – basically this is the bond-world equivalent of the bank that kept track of who had what deposit, and provided a way of transferring money between deposit accounts. Bearer bonds are more interesting – they belong to whoever physically holds the bond document and therefore are anonymous. That owner can transfer the bond to another by simply giving the transferee the document. Bearer bonds used to have little pieces of paper attached to them, called coupons, that one tore off and took to the bank, where one could get cash in return for the coupon. So bearer bonds actually look a lot like cash. They are mostly extinct now, since they were often used for tax avoidance or evasion. Also, if one misplaced a bearer bond, the wealth that it represented (for them) was often lost, disappeared, gonzo…

Bonds as a class have a particular relevance for blockchain – the ‘system of beliefs’ that govern how they work are mostly contained within the bond document itself, in hundreds of pages of legal language specifying how the bonds are to be transferred, what happens if the bondholder defaults on their payments, which courts are allowed to resolve disputes between bondholders and issuers, and so forth. However, in order to enforce contractual terms or resolve a dispute, a bondholder needs to approach the issuer or chase down the custodian/trustee, and if that doesn’t work, sue everyone – all of which is a time-consuming and expensive process.

Bitcoin, the Blockchain, and Digital Money

After that lengthy preamble, we can perhaps see, in the proper perspective, what bitcoin and the blockchain might offer to various societal stakeholders. Firstly, terminology: bitcoin and the blockchain are easily conflated, but for the purposes of this essay, bitcoin is a ‘digital currency’, while blockchain refers to a complex of underlying technologies that support and enable bitcoin and other digital currencies. Blockchain-based platforms may be available to everyone (public) or only accessible by a specific group of users (private). Bitcoin has attracted a great deal of attention and notoriety, but I, mirroring the FF/ACFL conference, will concentrate on the blockchain, which also happens to be the focus of intense investment by technology firms, venture capitalists, banks, and regulators. Furthermore, I will wave my hands and abstract away exactly what the blockchain is. Suffice it to say that it replaces the central ledger or repository variously referred to above, with a distributed ledger which is held at various nodes in a network, in multiple, synchronised copies. Every time a transaction happens, this information about asset ownership is updated across the network (essentially) simultaneously. Most of the complexity arises in authenticating transactions and in working out the time order in which transactions have happened, to avoid double counting/spending, ensure anonymity, reward the various agents in the network for participating, etc.  An engaging lay explanation of bitcoin (blockchain is easier to grasp if one starts with bitcoin) is here.  A sophisticated, but not especially mathematical, explanation of the blockchain, which starts with bitcoin’s implementation of the blockchain, is here.  For those interested in the undressed guts, see the Nakamoto PDF below which references Hashcash and other bits of prior work in this area.  Lastly, I’d be remiss in not pointing to the blog of he whom some, depending on their cultural background, might deem the Pretender, the false Dimitry, or a mere Anti-Pope: self-outed as Satoshi Nakamoto.

Smart Contracts

One of the most exciting, from a commercial perspective anyway, things about the blockchain is the idea of a ‘smart contract’. As I touched on above, bonds have fixed contractual provisions but in order to enforce any of them, one needs go to court or to a custodian. What if there was a bond which would automatically do certain things, like pay its coupon, or shorten its maturity date, in such a way that the (small) bondholder didn’t need to go fight the (big) issuer in court? A real-world example arises with Everledger diamonds that have blockchain-derived serial numbers etched onto them, accompanied by smart contracts that describe their provenance. One of the important points to note about blockchain, posited early on by bitcoin’s (ex-)pseudonymous inventor Satoshi Nakamoto, is the idea of immutability – once a transaction is inserted into the distributed ledger, it is impossible, or unfeasibly hard, to reverse. This means, in a contractual, financial, and legal sense, the history and provenance of diamonds (say Angolan or stolen) cannot be re-written or laundered. Another of Nakamoto’s principal aspirations was that, in a manner analogous to authenticating cash, authentication and verification of blockchain-based transactions should be something that could be done easily by an agent, even if they weren’t technologists or didn’t have the resources of a network node (i.e. a farm of computers mining bitcoins).

Devolving Funding, Ownership, and Contract

Smart contracts may also allow groups of (not necessarily rich or especially knowledgeable) collectors to jointly buy stakes in an artwork that none of them could have afforded singly. A digital contract would accompany the artwork, and the collectors would be immutably bound to that artwork. Conversely, the artist too could be immutably bound to the work, so that, in 30 years, when it sold at auction for millions, the artist could get paid. This is in contrast to the current situation, where artists or early-stage collectors don’t always benefit from the eye-popping gains in the value of artwork they created or nurtured.

It is however, important to point out that the issue I highlight above is not essentially technological – it is a contractual problem. Nothing actually prevents groups of collectors buying shares in paintings, just as nothing prevents artists from selling artwork with a restrictive covenant that constrains further sales or attaches to such  sales (and plenty of artists have explored these ideas of contract-in-art). The practice isn’t widespread presumably because few collectors would buy an emerging artist’s work with such a proviso. Having blockchain technology doesn’t really change the picture, in my view. What blockchain does do is automates execution of the relevant provision, and in the event there is a breach or dispute, might lower the cost of enforcement. Because smart contracts don’t require lawyers or court enforcement, and are almost instantaneous to execute, they become useful for relatively small sums of money. The diamond example above can easily be ported to art: Ascribe and MONEGRAPH provide content management suites that use blockchain to manage rights to digital art assets.  The notion of lowering barriers-to-entry is one of the most prosaic, and therefore most adoptable, applications of blockchain.

A high-profile applications in the area of funding and rights management was by the musician Imogen Heap, who has established Mycelia, a music storage and rights-management system built upon the Ethereum platform. Mycelia hosts an artist’s tracks, automatically updating as and if new and improved recordings are added; the information is stored in the cloud, and as a track is played, a smart contract is triggered to automatically pay the artist. Mycelia would also store other digital assets, such as liner notes, videos, etc., and contain provisions to allow derivative use: content to be embedded into other media (such as videos, advertisements, etc.) would trigger automatic payment to the artist. In theory, this all happens in the music industry already, but I suppose the point is that middlemen or artists’ agents handle the process, often badly (from artists’ perspective) and take a significant slice of the economics. In Heap’s words, Mycelia’s promise is that it might make the contracts between music artists and whoever ultimately is paying for content, whether record labels, advertisers or the public, much more transparent, easier to understand, self-executing, and thus, cut out layers of middlemen, while ensuring artists received as much revenue as possible, as quickly as possible.

Collaboration

It appears that Heap’s ideological perspective is that audience, listeners, can and almost should be brought in as partners with the artists, and thus Mycelia allows for more direct links between the two. An analogue of this forms the curatorial practice of Helen Kaplinsky, who talked about how to share assets in the current British cultural climate, where the concept of ‘civic publicness’ is crumbling, throwing into crisis the Enlightenment-era model of the museum as a public good. She has been converting a cultural space from a single-landlord entity into one that is owned by a community land trust. In a related project, she wanted to think about how artworks, accessioned into institutions but not currently on display, may still remain in circulation (and visible) rather than just sitting in a warehouse. In her vision, viewers, artists, and institutions would share an artwork in a fundamentally different way than the previous institutional model, which was binary – either a work was on-display or in-storage. More generally, the project tries to address questions that are coming up now in the contemporary economy & in art, revolving around what to do with artwork that might not have a ‘permanent’ or ‘stable’ state, something that has become key in museum conservation departments globally. Blockchain-based platforms like Ascribe provide a way of keeping track of interactions (ownership, borrows, licensing, insurance, liability, etc.), and because of the relatively low legal/financial cost, perhaps enable large-scale sharing of an artwork or a collection. Again, incorporating blockchain seems to me an incremental, and possibly worthwhile, technological improvement to an existing real-world solution: at MIT, students are already able to borrow artwork for their residence rooms.

Geo-wallets: Monetising the Self

Another intriguing applications of the blockchain lies in the integration of money, via the digital wallet, with GPS technology. Max Dovey presented a project undertaken with University of Edinburgh’s researcher Chris Speed where geo-location technology was connected with a digital wallet, such that the account balances within the wallet would change depending on how the user/viewer moved through Edinburgh. Another project called Handfastr also meshed geo-location with digital money, allowing couples to form temporary fake ‘marriages’ and temporary pooled accounts based on proximity: as long as the couple were near each other, they would have a joint digital wallet, and when they moved apart (spatially) they would revert to individual wallets.

These projects exemplify how small collaborative groups, physically co-located, can self-organise, and pool their financial resources easily and with a minimum of legal and contractual overhead.

In my view the more interesting perspective is an ambiguity of the whole thing: by linking money so closely to other digital technologies (GPS today, and in the future, say, bio-medical data from an Apple Watch), the move towards monetising individual identities and bodies continues apace. At a basic level, one can imagine large corporations, and the state, having even more information and potentially, control, over people. Given the enthusiasm with which people have adopted cashless, non-anonymous, immutable-record and friction-free payment methods (tapping debit cards, Apple Pay, or Uber), I can imagine many people will happily opt-in to linking more and more elements of their identity into a variety of surveillable and monetisable, albeit ‘free’ and convenient, networks.

From Self-Executing to Self-Replicating

Most of the approaches above have looked at funding, ownership, and sharing implications of the blockchain. Another potential artistic trajectory could be investigating the underlying technology itself. For instance, Plantoid is a system implemented on the Ethereum blockchain, that takes the initial form of a sculpture; however, once a certain trigger, described in an Ethereum smart contract, is reached, the blockchain instructs human collaborators to make a new version of the sculpture (and an algorithm embedded in the smart contract provides the specific dimensions of the sculpture to the human producers). The auto-generative potential, even if it is, at the moment, human-assisted, is combined with an economic metric, as encoded in the smart contract. If one takes a dystopian view, the project, loosely paralleling certain concerns about hypothetical self-replicating AIs, perhaps shows a future where mechanical things make themselves over and over, with human beings as mere workers or artisans operating under the (benevolent?) guidance of the an automated, financialised network.

Questioning Some Shibboleths: Доверяй, но проверяй

In the FF/ACF discussion, the word ‘trust’ kept coming up.  All the while, I was reminded of U.S. President Reagan’s little joke phrase he’d trot out regularly at his meetings with General Secretary Gorbachev: “Trust, but verify”.

In Nakamoto’s original technical paper, in which he drew up the bitcoin idea, he quite explicitly describes it as a system not predicated on trust. Specifically, it doesn’t require participants to trust the keeper, whether governmental or corporate, of the central ledger, as was the case with bank deposits or bonds. In fact, the technology makes distinctly non-idealistic assumptions: rational actors, often intensely individualistic even quasi-autistic, who don’t know each other, thus have no reason to trust each other or any third party.  Hence the importance of ensuring that each transaction can be authenticated and verified independently and quickly.

Other words came up in our discussion: ’emancipatory’, ‘radical’, which appear to be sentiments that come in from the hype around bitcoin. A cursory Google search tells us bitcoin and other crypto-currencies would eliminate the need for governments, central banks, courts, international law. Many of these aspirations seemed to gloss over the point that bitcoin doesn’t operate in a vacuum, it needs telecom networks and lots of computers, and in most cases, it interacts with real world laws and regulations. These quasi-public goods represent a societal subsidy (or shackle), often by profit-maximising evil capitalists and neoliberal-infected governments. Moreover, even self-executing ‘smart contracts’, pace theory, are unlikely to do away with disputes or fraud. What about when the system makes a mistake or simply breaks? Just as in the real world, we keep a wary eye on cash wallets, in the digital world, there will presumably be a need for users to constantly monitor digital balances, as well as familiarise themselves with applicable Terms & Conditions (which most of us currently Ignore & Accept). When something bad happens, counter-parties may, in extremis, need to resort to boring physical-world things like courts, lawyers, and contracts. As an aside though, Vitalik Buterin, the main force behind Ethereum, has proposed a ‘decentralised court’ that would arbitrate disputes.

My view, and I think that of some in the FF/ACFL room, was that the blockchain technology isn’t inherently emancipatory, just as it isn’t inherently repressive. As Vinay Gupta pointed out, blockchain can be used to support pretty much any political outlook. In fact, the reason banks are interested in it is pretty tedious stuff: it might dramatically shorten the amount of time it takes to finalise transactions in bonds, stocks, and other financial instruments. It might also minimise the risk of double-counting money or assets during the payment settlement period, as mentioned above. Similarly, governments are interested in blockchain because it might reduce tax fraud, make for a safer financial system, better track eligibility for welfare support, and increase employment.

Steve Fletcher at Carroll/Fletcher Gallery noted the ecological cost associated with implementing blockchain – again, hardly something that fits comfortably in the radical story-line. The ecological cost arises from the electricity utilisation of thousands of computers that, collectively, form the network which underpins the distributed ledger. These computers are constantly running to process transactions, a process sometimes called ‘mining’ (analogy with mining gold), and this eats up lots of power.

At a poetic level, insofar as economics can be looked at poetically, this process, called ‘proof-of-work’ in the blockchain jargon, is a loose analogy to what I originally defined money as: a socialised claim on labour (or assets). Hence, it would almost be philosophically perverse to expect a form of money, digital or physical, that did not perform this almost-sacramental act of abstracting labour into monetary value.

From another perspective, and this lies beyond the scope of this post, I’m also not sure how scalable the various blockchain technologies are – if everyone in the world used blockchain for all transactions, does that imply a huge increase in mining processes (with consequent ecological cost), or can a very limited stream of proofs-of-work be used to support all transactions? While blockchain might reduce the need for trust (in government or big corporations), it does so tortuously and expensively; I can’t help sensing a vague analogy with American survivalists who flee the Federal government for a rural homestead, armed-to-the-teeth and stocked with tinned provisions, bleakly awaiting the End of Days.

In summary

The FF/ACFL programme was a great opportunity to bounce ideas around, without a tightly-framed artistic, commercial, or ideological agenda. What I’ve tried to do here is sketch out how one might think about money generally, using the metrics of authentication, anonymity, convenience, fungibility, and contract, and apply these metrics to the bitcoin/blockchain complex. What seems to emerge is a potential for blockchain to devolve mechanisms and processes for funding for artists, as well as allowing various players in the arts ecosystem – artists, collectors, viewers, curators, and others – to define how they want to interact, with the possibility that sharing and artwork almost merge, or at least become as two sides of the same coin. Both of these potentialities seem more evolutionary than revolutionary. Other approaches offer fertile ground for exploring bio-politics and auto-generative physical artworks. A constant I observed through the entire conference was the inherent ambiguity of the technology: in contrast to the original libertarian or revolutionary claims made for bitcoin, the evolution of the technology today seems to offer as many risks of a dystopian future as emancipatory opportunities.

U. Kanad Chakrabarti  is an artist and former banker based in New York and London.

 

A Few Questions on Stranded Assets

Now for a post that’s a bit different from the usual rants about food or art:

‘Stranded Assets’ are a somewhat emotive topic that have moved from the relative backwater of Guardian-readership into mainstream investing thought.  The issue has been simmering away since about 2011, entwined with various fossil-fuel divestment campaigns, and headline-grabbing art performances.  Meanwhile, relatively quietly and in the more prosaic world of investment management, coalitions of investors, purely out of gimlet-eyed self-interest, have been pressuring the corporate world to take the threat of climate change more seriously.  With the 2015 COP21 climate-change conference in Paris, and Mark Carney’s widely-publicised intervention, the issue moved onto the front-pages.  We wanted to ask a few questions to assess how this issue might evolve, and how much, if any, is already priced into oil & gas stocks.

Is this a fringe issue or a real risk to investors?

Some investors, oil companies, and media have tried to paint the issue as something dreamt up by coddled, rich-world tree-huggers that has no relevance to actual portfolio allocation decisions.  While at one point this might have been true, it is no longer a gap-year fringe issue.  Look at BOE Governor Carney’s letter to insurers delivered at Lloyd’s in September 2015,  an HSBC research report from April 2015 on stranded assets, or Michael Bloomberg’s chairmanship of the Financial Stability Board (FSB) disclosure panel on climate-related financial risks.

What exactly is the financial risk?

There are at least 3 components to the financial risk to an investment portfolio.  Firstly, to the extent climate change causes substantial, secular changes to the global economy, most obviously a higher frequency of natural disasters (hurricanes, flooding, etc.), this could have implications for insurers’ liabilities as well as for broader portfolios with significant EM concentrations.  Much of these risks are concentrated in the significant equity/debt markets of Latin America and South/Southeast Asia.  So we are talking increasingly-frequent and large natural-disaster-related losses distributed across broad-market portfolios.

Secondly, if one believes that the 2° C global-warming target is something that, from a regulatory perspective, is likely to be enforced – meaning that most global regulators are likely to comply with their COP21 commitments to bring down emissions to a level consistent with this target – then it would appear that much of the world’s coal, oil, and gas reserves would be un-burnable.  Therefore, the reasoning goes, because the traded energy groups are partially valued using their hydrocarbon reserves, this would have significant negative implications for their market valuations.  In theory, if this regulatory response happened overnight, there would be a massive drop in energy group stocks, associated drops in utility shares, as well as companies that support the energy complex (pipes, pumps, etc.).  Given these sectors’ contribution (Vanguard’s all-world VT ETF is 13.7% invested in energy, basic materials, and utilities), there would be implications for broad indices, and therefore pensions, endowments.  Insurance companies, already exposed above on their liability side, may also run risks on assets.

Obviously this isn’t going to happen overnight. There are other caveats: technology to lower carbon emissions might improve – though, carbon capture & storage (‘CCS’) doesn’t appear to be going anywhere fast.  Also, much of the world’s oil & gas is owned by national oil companies (‘NOCs’) who aren’t investible anyway and operate in countries, such as Iran, Iraq and Saudi Arabia, where the regulatory response to COP21 may be quite different from that in the US/EU.

Lastly, carbon risk is tied up with oil-price risk.  At current prices, much of the reserves of certain oil & gas companies is uneconomical to exploit – Russian Arctic oil, Canadian tar sands, shale all need prices between $80-120 range to be profitable.  These assets can be said to be, to a greater or lesser extent, economically stranded due to low oil prices.

How might this play out?

From a pragmatic perspective, the risk to portfolios lies over a 1-3 year horizon, as the regulatory response evolves.  The risk also lies in large institutional investors themselves placing pressure on companies, or indeed divesting, from fossil-fuel groups.  To the extent that investor coalitions, such as the IIGCC, which represents €13TN of assets, can gradually push energy companies to publish sensitivity analyses, at least the scale of the problem at an individual company level would become clearer.

At present, companies and equity analysts are able to assess the impact of prices on reserve valuation.  But it is much harder to disentangle the parallel effects of oil prices and climate-change regulation on the value of reserves.  Absent any other information, the base-case assumption might be that companies under pressure on carbon-emission regulation, would exploit their lowest-cost reserves first and mothball high-breakeven reserves.

In 2015/2016 there have been a number of high-profile divestments, particularly from coal groups, for instance by the Danish fund AP2, the Rockefeller Family Office, and the Norwegian Government Pension Fund.  There will undoubtedly be more – though it’s not clear that divestments, at least at the current rate, pose a significant portfolio threat to investors, since someone else buys those shares at the market price.

A game-changer could be if more SWFs joined Norway in either divesting or aggressively pushing for changes in company policies, not least because they are amongst the largest investors, with the most patient capital, and with potential access to infrastructure projects and assets that are not publicly traded.  Moreover, the biggest SWFs often belong to hydrocarbon exporters, who, one might imagine, would like to diversify their exposure away from energy.

China is another force to watch – in the 13th 5-year Plan, to be announced in March 2016 – there are expectations of significant initiatives to reduce carbon emissions and promote green technology.  The date often mentioned is 2030, when China’s carbon emissions are expected to peak, but until the Plan is actually announced it’s not clear what, if any, immediate steps are expected.

On the other hand, a more granular picture on regulatory response scenarios is provided in a paper from the Grantham Research Institute.  The interesting bit of it is as suspected – in what sense is it fair to ask EM countries, significantly poorer per-capita than the DMs, to bear the inevitable costs, howsoever pro-rated, of transitioning to a low-carbon environment?  Moreover, how reasonable is it to expect that countries who have short or non-existent histories of civil society/institutions, a record of policy reversals, endemic corruption, and weak states, will embark on, and deliver on, COP21 commitments that are intensely challenging, from institutional, legal, political, financial, and technological perspectives.  The report, euphemistically, highlights Argentina, China, India, Indonesia, Saudi Arabia as having ‘potential for increasing support [for implementing COP21 commitments]’.  Lest our SUV-driving North American friends pat themselves on the back, neither the U.S. nor Canada do much better in the report.  So it’s quite possible that, from a regulatory perspective, very little happens in the next few years; though that would seem to imply that the necessary adjustment to policies, and therefore investment portfolios, when it inevitably comes, will be that much more severe.

How have the oil & gas companies responded?

As stated above, the quality of disclosure and sensitivity analyses is presently most inadequate.  Even so, European majors score better – for instance, Shell, Total, Repsol, GALP all score highly on the Carbon Disclosure Project’s rankings. Chevron and Exxon Mobil are on record as being resistant to shareholder resolutions proposed in 2015.  However, Exxon Mobil did produce a climate-change disclosure report in 2014, albeit without any meaningful sensitivity analysis.  More specific information on company responses, including mining groups, can be found at the Association of Chartered Certified Accountants.

Exxon Mobil’s 2016 energy outlook, which covers the period out to 2040, provided their take on how carbon reduction commitments might be met for the system as a whole, namely through dramatic reductions in carbon-dioxide and energy intensities (50% and 40% respectively).  According to the FT, this stretches credibility given apparent lack of political will, certainly in the US and even in the EU, as well as an inadequately high carbon price.  Looking at the price of American gasoline and the alacrity with which people drive in most of the U.S., outside Manhattan anyway, I can see the pink paper’s point.

Lastly, some companies, such as Total SA have invested in renewable energy as a hedge, howsoever small, to their extractive activities.

Is the risk already priced in?

At current oil prices (Brent future for December 2018 delivery is $46), the market recognises that certain high-breakeven reserves, such as in the Arctic or in Canadian tar sands, are unlikely to be monetisable.  Thus, economic stranding is reflected in share prices of groups and funds that are highly exposed to these reservoirs.  It is much less likely that the risk is priced in for the integrated majors, given their diverse properties and mid- and down-stream operations.  It is even less likely to be priced in for utilities and infrastructure companies.  Again, there doesn’t seem to be much analysis or disclosure around the (presumably) differing sensitivities to economic and carbon-related stranding, in part because the regulatory regime that would drive this is yet to be announced.

How to play this?

At the moment, this is basically a structural risk that’s relatively hard to avoid – since in theory everyone from energy companies to utilities to car-makers are somewhat exposed.  Moreover, energy groups are steady dividend payers, for now anyway, so dropping them has a significant financial cost.  Thirdly, coal is mined by integrated miners who also pull other stuff out of the ground, so divesting coal might mean losing much of the metals & minerals complex.

The simplest approach, for a small or passive investor if not the system as a whole, is to remain invested in a broad index, and hope for the best.  There are two reasons for this: one is that, depending on how the situation plays out, portfolio impacts could vary tremendously.  Mercer Consulting highlight drastically different return and composition-of-return scenarios depending on whether regulatory and investor actions prompt rapid, proactive and coordinated, as opposed to, delayed and fragmented, responses.  Thus, this risk could be viewed as an exogenous factor, a bit like catastrophe, war, or un-planned death, that cannot be diversified away simply or costlessly.

The ‘dumb’ strategy above can be improved by removing coal, which no one has anything good to say about (not even Goldman Sachs), and (possibly) by removing high-breakeven companies, such as those involved in Canadian sands or US shale.  Whether one then tilts the portfolio from oil to the relatively cleaner gas becomes a judgement call.  Further tilts are possible away from E&P towards pipelines, infrastructure, refining, or distribution, again probably with diminishing returns (in terms of yield/concentration trade-off).

lowc_crbn_vt_perf
Vanguard VT (all-world ETF) against two ‘low-carbon’ ETFs Source: E*Trade

Some new ETFs, including CRBN from iShares and LOWC from State Street/SPDR follow indices with lower carbon footprints, such as the MSCI ACWI Low Carbon Target Index, jiggling other index constituents to achieve similar dividend yields and correlation with the much broader all-world MSCI ACWI Index.  This chart above shows performance relative to Vanguard’s all-world ETF VT, albeit over the slightly meaningless timeframe of a year.  These ETFs use company disclosures on emissions to re-weight their allocations, but it’s not entirely transparent how meaningfully exposure to carbon regulation has actually been reduced: for instance, Exxon Mobil and Chevron are still in the portfolios, albeit at tiny allocations, while Enbridge Inc, a gas-pipeline operator and Ultrapar, a Brazilian conglomerate with downstream operations represent meaningful holdings in LOWC, while not showing up at all in VT.

Screen Shot 2016-02-29 at 12.59.55
Clean energy ETFs compared to VT (Vanguard all-world) and VDE (Vanguard energy) ETFs on 5-yr horizon

A complement to the approaches above could be investing in clean, renewable, alternative energy and related companies.  There are a number of ETFs in the space (ICLN, PZD, QCLN, FAN, NLR, KWT, YLCO, GEX).  Most of them look like dogs: over 5 years they’ve substantially under-performed the broad market (VT) as well as energy (say VDE) ETFs.  Fukushima Dai-ichi wiped out the uranium ETF.  Over 3 years the others seem to have gone through a bubble as oil prices went up, making their relatively-expensive technologies look viable.  Then, in the last year as oil prices have crashed, substitution effect has made green technologies again look uneconomic.  So unless one believes substantial subsidies for green tech are forthcoming, that oil prices will shoot back up, or that regulatory and public attitudes towards nuclear will change, these feel a lot like a levered play on oil prices, where investment timing becomes critical.  However, within this universe, specific plays might be interesting (such as measurement companies or water-treatment companies), due to the diversity of their businesses and/or relative independence from subsidies and oil prices.

Screen Shot 2016-02-29 at 13.00.09
The clean-energy bubble (2013-2016)
Screen Shot 2016-02-29 at 13.00.31
Clean energy follows oil down in 2015-2016