13 October 2016
Would a flash-crash dash be rash?
Fat fingers, algorithms, skullduggery or a portent of doom?
by Frank O’Nomics
“The sky is falling!” shouted Chicken-Licken. “So sell Sky” responded Foxy-Loxy. That is the kind of response that would create a flash-crash; a concept which is back in the news after a dramatic short-term fall in the value of sterling last week, a fall which highlighted concerns about the potential economic consequences for the UK of a hard Brexit as well as calling into question the efficiency of global markets. In the space of a few minutes in the early hours of Asian trading last Thursday, the £/USD exchange rate fell over 6% , adding to a fall of over 15%, to a 31 year low, since the referendum vote in June. While the most of the fall was not sustained, the event has highlighted the vulnerability and jittery nature of markets during a period when we still do not know the terms of our departure from the European Union, and is being used try to influence the Brexit negotiators into a softer stance. However, a flash-crash can occur for a number of reasons. Sometimes the move is the result of an unfortunate combination of factors, some of which may be artificially created by devious traders for short-term gains, pushing the level of an asset class to an artificially low level. The move in US stocks supposedly prompted by a trader operating from a semi in Hounslow is such an example, and moves like this do nothing more than create a nice opportunity, at least for an innocent by-stander, to buy assets on the cheap.
However, sometimes a flash-crash can give an indication of where an asset class is heading over a longer period – occuring when liquidity disappears and markets have to very quickly find a new clearing level where buyers and sellers are balanced. The question regarding last week’s move is, what kind of flash-crash was it? We need to look at what triggered the flash-crash, because this can give some idea of how sensitive the markets are to Brexit negotiations, as well giving pointers to how bad the situation could get.
First of all we should ask: Why does any of this matter? The sharp move down in sterling after a period of moderate calm has been likened to Wile e Coyote, having run off a cliff realising that he has an anvil attached to his leg. Markets don’t always move smoothly from one point to the next, with discrete sudden shifts often needed. The move in sterling can be seen as a the result of a further evaluation of the consequences of Brexit, this time in response to the “hard” Brexit soundbites from speeches at the Tory Party Conference. A steep move may just get us more quickly to a point that we were going to reach at some stage anyway. Many commentators have pointed out the negative economic consequences of the move in that, as we spend around 30% of national income on imports, a 20% fall in sterling reduces our effective wealth by around 6%. However, the longer-term impact of making UK exports significantly more competitive, and the help the inflationary consequence gives the MPC in achieving its inflation target, may be seen as more positive. Recent correspondence between Mark Carney and Philip Hammond suggests that neither party is particularly concerned by the decline of sterling, and a former IMF chief, Ashok Mody, has suggested that Brexit has created a healthy correction of a 20-25% overvaluation.
Suggestions that the move was all down to Brexit have some validity. Sterling had closed at $1.26 on the evening of 29th September, down 3 ½ cents on the week, following successive government speeches that led markets to conclude that a hard Brexit was in prospect. In the early hours of Thursday morning the FT released a story revealing Francois Hollande’s tough stance on Brexit negotiations and this coincided with the fall in sterling developing a new pace. However, it is hard to argue that it was news flow that prompted the flash-crash. The Hollande story was released during the fall – it did not precede it, and hence could only have caused the move if someone knew it was about to come out. There is then a vague smell of some skullduggery, but nobody has found any evidence of this (at least not yet, although Mark Carney has asked the BIS to investigate), so for now we should regard the story as having compounded rather than prompted the move. If it wasn’t news on Brexit then, how did it happen?
The standard response is to look for a “fat finger”, where a trader has, at precisely the wrong moment, submitted a trade that is either in the wrong direction (buying when he should be selling) or more likely the wrong size (billions instead of millions). Again there is little evidence for this. If a trading desk has suffered a significant loss it tends to be known about very quickly – not least because there are usually efforts to cancel trades. The other widely discussed potential trigger is the activities of algorithmic trading funds. These automated funds may have kicked in when a certain key level (probably $1.24) was broken and, due to a lack of liquidity and more robotic funds acting because of a momentum trigger, the reaction became extreme. This explanation carries more weight. The crash occurred when liquidity was very poor, when only Asian based banks were trading, and few of them would have been active at 7am. It was also a day when many Chinese traders would have been observing a bank holiday and few banks would have wanted a position ahead of the US payrolls data and a long weekend in a number of Asian countries. As a result, when a reasonably large selling order was seen, no one wanted to take on a position and a rapid game of pass-the-parcel ensued.
It may be that the move demonstrates a “liquidity illusion” in markets where, when tested, there is little substance. In fact the flash-crash could actually improve liquidity by helping to create bids below the current market level by those not wanting to miss the opportunity to buy as cheaply as others did last Thursday.
If Thursday’s flash-crash was little to do with economic fundamentals then perhaps we should regard it as nothing more than a missed trading opportunity. If we look at the OECD data on purchasing power parity, sterling has moved the cost of a basket of goods bought in the UK by someone with a dollar income from being around 30% more expensive just over a year ago to a point of near equivalence now. This would suggest that we are near a good support level for sterling, with little fundamental justification for sub-$1.20. Sadly, purchasing power parity theory is just that – a theory, with as much relevance as a stopped clock, which is correct twice a day. A decline in sentiment could push the pound well through a level that is regarded as cheap and there are a lot of hedge funds who are betting on just that, judging by the evidence of a sharp increase in the number of “short” positions. While anyone who bought £/$ at 1.15 would have been feeling very clever by the end of the day, that profit has been steadily eroded, with yesterday’s story of a £66 billion potential cost of a hard Brexit causing a move towards $1.22. The ultimate level at which people are prepared to hold sterling in light of Brexit risks may be closer to $1.10 than to $1.25. Reverting to the earlier analysis, this points to a significant short-term erosion in national income and a longer-term rise in inflation. Foxy-Loxy may be right after all.
If you enjoyed this article please share it using the buttons above.
Please click here if you would like a weekly email on publication of the ShawSheet
