Imagine an experiment where you were able to shut down one of the largest centers of commerce in the world for a few days unexpectedly. What kind of impact would something like this have on other venues for trading around the globe? During the first week of Nov. 2012, this hypothetical turned all too real as the U.S. markets based in New York City were shut down on Monday Oct. 29th and Tuesday Oct. 30th. Our initial findings show that with Wall Street going offline, the Toronto Stock Exchange (TSX) was profoundly impacted when looking at the raw number of quotes and trades, but with a corrisponding increase in average shares per trade, the actual number of shares traded and hence economic imact wasn't as large as it might appear at first blush.
*Images courtesy of wikipedia.com
When Wall Street shuts down, it's a big deal. The lifeblood of trading that public corporations depend upon for extra capital to grow and expand grinds to a halt. Since the U.S. markets are such a major player world wide, it's not surprising that there would be some effect on other markets. But when it's quantified, it really opens your eyes into exactly how interconnected the world's markets have become.
Number of Quotes
One of the first things we looked at was the raw number of quotes, or orders. As trading has evolved from being human to human via the market floor or telephone to being computer-to-computer, the amount of quotes that are generated is a good sign of computer related activity. Here we show the number of quotes the week preceding Hurricane Sandy and the two days that U.S. markets were shut down.
Number Of Shares
While the number of quotes dropped by 70%, the number of actual shares traded only dropped by roughly half that ammount with a 34% decline. This indicates that trading was actually relativly strong. The real thing that suffered when Wall Street was shut down was the volume of quotes commonly associated with computer algorithms. When we look at the shares per trade ratio a little bit further down, it will become evident why the raw number of shares didn't drop as much.
Once you have the data for both quotes and trades, you can calculate the quote to trade ratio, a great indicator of how much "noise" is going on in the markets. If there are many times more quotes than trades, it might be an indication that there is a lot of HFT or computer algorithms participating in actions. Everything in moderation. We here at Nanex are all about technology and speed, but when systems are pushed to their breaking point, things fall apart rapidly. A prime example of high quote to trade ratio was discussed in one of yesterday's reports on the DoS Algo and its ability to all but halt trading in whatever security it was running on. When thinking about high quote to trade ratios, a general rule of thumb is: the more quotes, the more computers. Some computer models want the least volatile, most predictable trading environment possible so in the advent of an unknown event like U.S. markets going dark, they may have just followed the herd and stayed offline until everything was back to normal.
The last indicator we will look at is another indicator in how much electronic trading is going on in the markets: shares per trade. As the markets have become faster, the number of shares per trade has dropped. Part of this is due to the retail investor's relatively small amount of capital investment, and part of it is due to the need for institutional investors to chop up their large amount of required shares needed into smaller chunks to try and avoid detection by predatory computer algorithms. This is where it gets interesting. The previous three charts resulted in a net decrease, but the shares to trade ratio actually INCREASED in the TSX exchange while the U.S markets were offline. We normally equate a higher share/trade ratio with there being more "real" trading going on. With the markets having so many participants pulled out, those parties who might want to move larger amounts of shares in a single trade might have felt safer doing so knowing that some of the typical machine based algorithms designed to detect and pick away at large orders might have been offline either by circumstance or choice.
Now that we have seen the broader overall market impact, lets drill down into the details of a few prime examples of how much the U.S. market shutdown played in some of Canada's largest companies: Barrick Gold (ABX) and Silver Wheaton (SLW)
Example 1: Barrick Gold (ABX)
One of the prime examples of this trading activity being shut down is in the security for Barrick Gold. Here are a few facts about Barrick:
One would think that this Toronto based, Canadian company would be trading strong and true on the TSX even when the U.S. markets are down.
Let's have a look at the data shall we?
Number of Quotes
As you can see, the U.S markets being closed affected Barrick Gold (ABX) in a major way. But, not as much as our next example.
Example 2: Silver Wheaton (SLW)
Here are a few interesting facts about Silver Wheaton:
Silver Wheaton (SLW) looked like a ghost town when Wall Street closed the swinging doors to the markets. SLW quotes dropped by 97% and it's trades dropped by 85%!
Number of Quotes
Wow! 97% less quotes and 85% less trades when the U.S. market wasn't open. This is the poster child in what kind of influence shutting down of the U.S. equity markets has on certain securities in the TSX.
Obviously taking one of the world's largest markets offline for any reason will have an impact on today's increasingly globalized market structure. Even more so when it's not a planned event. Before this exercise, we were unsure exactly how much of an impact the U.S. shutdown had on TSX. With 70% drop in quotes, it was actually quite substantial. But if you take into account that the average number of shares traded only declined by only 34% due to an increase in the number of shares per trade, the actual impact to TSX in trading terms was less severe than if you just look at raw number of quotes and trades. As major storms like Katrina and Hurricane Sandy catch us by surprise, it also serves as a reminder that the U.S. markets have a ways to go in prepping for the next major disaster. We think this event has opened the discussion for redundant systems and measures we can take going forward to make it more resilient so that the algos in TSX don't get so lonely. Eh.