something.com

My previous post reviewed the potential arbitrage opportunity available over a USD transiting across CNY then over to a currency basket and finally back to USD transaction. The theory underpinning this is known as APT, or Arbitrage Price Theory, an important founding stone to EMT, or Efficient Market Theory.

Arbitrage Goes Round and Round, Round and Round

In theory, in an efficient market, all arbitrage opportunities, that is trades that present a ‘riskless’ return, should not exist as prices already present all information available to the market and therefore already have all inefficiencies priced out due to their potential arbitrage value. If you’ve noted the circular logic then you’ve understood, if not, read again.

Practical Arbitrage

The reality of APT is not quite as a straightforward. There are a number of factors that may act as systemic providers of inefficiencies in markets and can therefore cause arbitrage opportunities to remain in the market with limitations preventing their use and subsequent ‘pricing out’.

But Do These Limitations Necessarily Prevent Arbitrage?

No, not necessarily. In theory, regulatory arbitrage should in itself provide sufficient incentives to ensure that markets align and liberalise across sufficiently well enough in order to overcome the systemic obstacles to trade-related arbitrage. As I said, that’s in theory, in practice of course, there is ‘friction’ if you will.

To take the previous article as an example and to name but a few aspects to the trade that would pose problems in practice (I’ll list below some solutions to the problems raised here):

    1. Decimalisation of trades: FX operates in ‘lots’ usually of a 100k of the base currency which doesn’t necessarily mean that you can further ‘trim’ the lot when it comes over to the next trade. The example is perhaps most explicit with cross-market equity arbitrages: you can not ‘usually’ split up a share just because you would like to trade point two or so of a share.
    2. Barriers to Entry: Markets are not, unfortunately, as uniformly accessible around the world as we would like. Though progress in this area will continue, there still remains a lot to be done on this front. For FX transactions, this becomes a particularly tense aspect of the trade as the physical SPOT CNY currency market is both highly regulated and controlled.
    3. Bid/Ask Spreads: Depending on where you sit in the market, the transaction fees set into a trade can negate gains available through arbitrage. You will notice, in my past example, however, that all the currencies were already quoted with the price rate. So assuming a consistent spread, they would affect all the trades uniformly. But again, a consistent spread is a big assumption.
    4. Operational Risk: This is a big one in almost any commercial venture, let alone FX operations. For the FX arbitrage example listed on the previous page, quote ‘slippage’ can pose a significant cost on any ongoing operation. Examples of ‘fat fingers’ come to mind (was that 1 lot or 11 lots?), or sudden lock-outs over a trade leaving a position exposed over one cross. In the FX market where volatility is pretty much a given, this can have ridiculously high impacts on an end-of-day position.
    5. Uncategorised Tail Risks: Okay, so I’m kind of just renaming a whole bunch of other aspects here and lumping them in one category. But think political risk, environmental risks, the sun turning super-nova overnight. None of these are good for your book and, unfortunately, Murhpy’s law will tend to apply when and where it hurts the most. Some of the smartest people out there get caught out on this front (and I mean nobel-laureate styled smarts), the example of Long Term Capital Management comes to mind here.

How To Manage Your Exposure

Well, beyond the trade, this then becomes an issue of ‘management’. This is perhaps why in the ‘buy-side’ of the industry, fund executives are known as ‘investment managers’, ‘portfolio managers’ and the like. Basically, one needs to confront these risks with a commercial appreciation of their impact and therefore understand that an investment only becomes valuable once a risk-weighted value can be gained and secured with a high degree of confidence.

So returning to the issues mentioned above:

    Decimalisation: This issue is one confronted on an ongoing basis by financial markets. Increasingly, large market-makers actually do support further granularity in price and potential splitting of trade order parcels. Conversely, the very factor of decimilisation differences between markets can actually be ‘arbitraged’ itself, essentially through the use of scale and/or leverage. Of course, once leverage is in place, this does add on yet another factor that needs to be managed across (see Basis Risk in Foresight of the Curve and Inflation and the Curve). The rise of electronic trading venues and so-called dark pools or high-frequency traders further assist in the decimalisation factors available in the market: if a market-maker uses scale to arbitrage out decimalised trade advantages then he just needs to ensure that he has a sufficient liquid book to collate and aggregate his positions over time to support his order flow, just as a quick example.
    Barriers to Entry: Right, this one is a bit controversial. Let me use a bit of an analogy here: a market has forbidden the trade of a particular good but does that necessarily kill the existence of a market of a particular good? No, so called ‘black-markets’ then tend to emerge. Now, let’s say that this ‘black-market’ can then be hypothetically transported ‘outside’ of the regulatory zone and basically provide an external venue for the transaction. At which point it is not ‘illegal’ anymore, just basically replicating, synthetically, what might have been happening within the regulated zone. Economically speaking, there shouldn’t really be much of a difference between the two assuming information and incentive patterns remain consistent across both the ‘restricted’ and ‘unrestricted’ venues. For foreign exchange, this practice appears in a number of ways, but to give but one example, there is the so-called NDF market: or Non-Deliverable Forwards. Without wanting to get into too much details here, these basically allow ‘price-discovery’ to occur over an economic exposure without an explicit holding of particular side in the exposure. Now, of course, the NDF market isn’t the FX Spot market either, it actually exists as a sub-component of the real market. But again, re-applying regulatory arbitrage over this should then, in theory, ‘re-align’ it back to the ‘real’ market.
    Bid/Ask spread: Transaction costs exists, and as viewed in the issues raised above, they can pile up and add sufficient uncertainties over price at any point in time. But again, assuming a highly liquid market, the trend for these would be to reduce and fall (in theory). More of a concern for any particular transaction, however, is the fact that this transaction cost factor can be uncertain. It’s one thing to model a fixed or constant proportion cost, but a stochastic cost with its own dynamics and underlying equilibrium factors renders the whole process, technically, unpredictable.
    Operational Risk: Operational control is an important facet of any financial operation. The ability to minimise, control and quantify operational performance is critical in order to maintain a viable and predictable enterprise. Clearly, this isn’t limited to finance, whether it be flying a plane, running a power center or managing the till at the corner store, operations can prove an important facet in the day-to-day management of any business. The key word here is management. This is a daily and ongoing reality which is never truly negated. More to the point, the very application of operational risk control measures can provide further opportunities of systemic risk entry when badly initiated or mismanaged. Operational risk control is not a ‘mission-killer’ nonetheless, instead it is a cost (albeit a fairly important one) that requires oversight and leadership.
    Uncategorized Tail Risks: There are known knowns, known unknowns, unknown knowns and unknown unknowns and then there’s the too darn remote risk to be able to categorise as either known or unknown. More to the point, some things just jump ship, from what was once in a contained and controlled position to the way, way, way out there level. These won’t disappear no matter what or how hard we try and confront them. I suppose the best solution in this instance is to always maintain a certain level of responsiveness-slack. In case something goes wrong, hit the any key, or call for reinforcements…

So Wait, Was There Arbitrage After All This?

Well, let’s put this way, 0.30% or 30 Basis Points is a lot. Like, a really, really, big lot here. Compound that per day for 250 business days in a year and you start to see the scale. Compound that for potentially a transaction rate that can match the liquidity out there per day then compound it for every business day and then you start to see how bad this is.

More worrying, is the fact that this arbitrage is almost constant. It would be one thing if it popped up every now and again, if it just randomly blipped up and then slipped away, but it is a bit too darn predictable and stable to just discount out. This isn’t good. Technically, in theory, this shouldn’t be here.

But Don’t All The Issues Mentioned Above Confront This?

The problem with the ‘issues’ mentioned above, is that the majority of them are actually fixed costs. In other words, once confronted, it’s just a question of then amortising the cost into the return of the trade.

Basically, this arbitrage shouldn’t be there. Now perhaps all my FX data is wrong, perhaps, there’s no time in a day when the trade can match up (though this could potentially be absorbed through adequate intraday position monitoring) but the biggest problem with this arbitrage is that it then skews everything else on its path.

In other words, once an arbitrage trade is identified, one needs not trade the actual ‘arbitrage’ itself, one can merely take positions over the ‘consequences’ of this arbitrage opportunity. Technically, this shouldn’t be happening, and yet it’s been going on for some time now.

Comments are closed.

Disclaimer: Material posted on 24-something does not contain (and should not be construed as containing) personal financial or investment advice or other recommendations. The information provided does not take into account your particular investment objectives, financial situation or investment needs. You should assess whether the information provided is appropriate to your particular investment objectives, financial situation and investment needs. You should do this before making an investment decision based on the material above. You can either make this assessment yourself or seek the assistance of an independent financial advisor. 24-Something, associated parties and Tariq Scherer accept no responsibility for any use that may be made of these comments and for any consequences that result.