Thursday, April 12, 2012

Trade parasites feeding at the heart of the ASX

By Alan Kohler

In the Australian Securities Exchange's Sydney data room, which is about the size of a big lounge room, there are six "cuckoos". These are the banks of servers installed by high-frequency traders.

They sit against the wall opposite the ASX servers and each is connected directly into the host by a fat fibre optic pipe. Each cable is precisely the same length by agreement with the ASX so that none gets an advantage; if one server is closer to the input, its cable is looped around to lengthen it.

Think about that: one less metre of optic fibre carrying data at 299.8 million metres per second - that is, the speed of light - would give one share trader an unfair advantage over the rest. It suggests that something pretty quick is going on.

The question is whether it's fair to the rest of us; whether those six parasites with their suckers fastened directly into the heart of the ASX should be allowed to get away with it.

The ASX is no longer a regulator, just a business, so it says that if the practice is legal and it pays a fee – not to mention a handy rent in the data room – then it can't and won't stop them.

For global regulators, it's actually too late: high-frequency trading accounts for as much as 70 per cent of the volume on American stock exchanges, including the NYSE; the time to control it was 10 years ago.

What do the computers and their algorithms do? Well, as my relatively low-frequency brain can understand it, these machines constantly monitor order flow into the ASX servers, and the sophisticated programs can pick up patterns that indicate when a reasonably large order has been placed. What they then do, in effect, is "front-run" – that is, they buy ahead of the order and make a small spread selling into it.

In other words, by operating at the speed of light they can "feel" a buy order coming and can dart in front of them and ensure that the buyers pay a little bit more than they were going to, without noticing a thing.

These operators begin each day owning no shares and end each day in the same position, but they make a lot of money by doing thousands of trades every day: it's a high-volume, low-margin business.

It's not known how much money the HFT traders make, but whatever it is, they weren't making it 10-15 years ago, and stock market returns have not gone up in that time, so whatever they make has come out of someone else's pocket.

That someone, of course, is you. The buy orders that the HFT operators are front running come from the superannuation funds in which ordinary people have their money. Now when they place an order, they usually end up paying a cent more than they would have because they are buying from someone who didn't own any of the shares 10 microseconds ago and only bought them to make that quick cent.

HFT represents less than 10 per cent of the volume of the ASX, but in the United States it is much more, and there is no reason to think we won't follow the US.

Should something be done to stop it? I think so, but it's too late.

HFT firms like the privately owned and aptly named Getco (Global Electronic Trading Company), the world's largest HFT operator, produce a large amount of self-justifying research material based around the proposition that they help investors by providing extra liquidity in the market.

This, plus presumably the hiring of expensive lobbyists, has snowed legislators and regulators and let the practice flourish, to the point where the parasites are taking over the host and it's too late to stop them.

Stock exchanges the world over are now making a fortune from renting space in their data rooms to high-frequency computerised traders and would probably collapse without it (the ASX would not – yet.)

As a result, investors are abandoning the "lit" markets and using "dark pools" instead. This simply refers to off-market share trading away from the official stock exchanges provided by investment banks where big investors know they are not being picked off by high-frequency front runners. The problem with that is that these "dark pools" are not properly regulated or transparent.

The joke is that in many cases, the same investment banks are doing both the high-frequency trading and running the dark pools; they are causing the problem and solving it, each for a handsome profit.


Tuesday, April 3, 2012

Gold Miner Ampella Hoists For Sale Sign…Reluctantly

Comments: Ampella just released two announcements related to their budget for drilling/exploration and their intention to open the company for possible takeover bids (For more info, click here for drilling targets and here for the takeover). The unreasonable and ridiculous sell-down of AMX share price could have led the management to announce these statements, which could give a more realistic picture to the public about the value this company holds. What to do. When investors in general don't know how to appreciate the value of this company and drive it down to ridiculously low levels, then just let the takeover bids do the talking. Every now and then, I saw some sellers purposely drove down the share price, even when the price was trending upwards. Could it be that some brokers purposely drive down the share price so much until a takeover bid is imminent? I don't know. But it's good that the company is taking steps to address and stem the slide in share price. Do note that major shareholders and directors had been holding the shares tightly even when the price was sliding from above $3 to current $1 level. This says very much about the confidence that they have in AMX. Since these announcements, the share price had risen 15%, but for me, this is only the beginning. There will be exciting times ahead for AMX. 

For previous posting, click here

By David Winning

West Africa-focused gold explorer Ampella Mining certainly looks a reluctant seller after firing the gun on a sales process in the wake of several new bid approaches.

The Perth-based miner has mandated Macquarie Capital and Gresham Advisory Partners to run a formal process after receiving interest from “several global players in the gold sector” on top of separate requests for due diligence made late last year.

The bids look opportunistic: Ampella’s shares have fallen from above A$3.30 at the end of 2010 to A$1.05 now, giving it a market value of 250 million Australian dollars (US$260 million). In contrast, the stock of ASX-listed peers that also have gold assets in Burkina Faso, such as Gryphon Minerals, aren’t down as much.

Ampella Chairman Peter Mansell pointedly stressed Monday that with around A$55 million dollars in cash on its balance sheet it would have to be a knockout offer to secure a board recommendation.

Its cash pile means the company can fund all its exploration and development work at the Batie West project in Burkina Faso for the next 12-18 months. In short, no firesale.

“With our strong balance sheet, Ampella is now very well placed to continue down its current path and in the absence of a compelling alternative that is what we intend to do,” Mr. Mansell said in a statement.

“Having said that, we have had approaches from serious global players in the gold sector and we concluded that we had a duty to all shareholders to properly test whether any one of those parties is prepared to make an offer to acquire Ampella on terms that would be attractive to our shareholders.”

After strengthening balance sheets when gold prices surged to a record high above US$1,900 a troy ounce last year, mining companies are now setting out on the acquisition trail.

ASX-listed Norton Gold Fields, which owns the Paddington gold mine in Western Australia state, halted trading in its shares at the end of last week after receiving an unsolicited buyout approach.

Earlier this year, a person familiar with the matter told Deal Journal Australia said Newmont Mining has received several expressions of interest in its 51% interest in the undeveloped McPhillamys deposit in Australia’s New South Wales state, which contains an estimated 3 million troy ounces of gold.

Ampella said its sales process is likely to take 2-3 months, and it wouldn’t be updating the market on the status of discussions with interested parties until it had made a final decision.