Go to the Globe and Mail homepage

Jump to main navigationJump to main content



Mining deeper to predict who will outshine Add to ...

The U.S. Federal Reserve's latest move to spur the economy with government asset purchases has sent prices of precious metals, and gold bullion in particular, to new heights.

The rise follows a similar pattern set after the Fed's first blast of quantitative easing last year, which had gold prices soaring more than 40 per cent. The increases this time will be less dramatic for several reasons, says Steven Green, an analyst with TD Securities Inc. Those include the fact that the central bank's move was highly anticipated this time around, gold and related stocks were not oversold in advance and QE2 is only one-third the size of the QE1 program.

But Mr. Green expects that we will see a "very positive environment for gold for the next several years" thanks to low interest rates, further rounds of quantitative easing and continued purchases of bullion by the world's central banks.

Obviously the bullish conditions won't translate into equal gains for the industry. Analysts use a number of financial metrics to try to separate the wheat from the chaff. They look at a company's production volumes, gold reserves and cost of extraction, commonly referred to as cash costs per ounce. They also consider cash flow per share, price to cash flow, net asset value per share, price to net asset value and debt to equity as important indicators.

In we have collected a set of data that can be layered on top of traditional industry metrics. It comes from StarMine Corp., a Thomson Reuters service that ranks analysts and gathers earnings estimate data to which it applies proprietary research to detect momentum and other factors.

"One could once make money by trading on earnings surprises and analyst estimate revisions. That's no longer the case. Today, you can't trade the event, you have to anticipate it. That's what these measures have proven effective at doing," says Tim Gaumer, director of fundamental research at Thomson Reuters.

The first measurement is called StarMine SmartEstimate, which is a proprietary blend of analysts' estimates that aims to more accurately forecast upcoming results than the consensus estimate. It gives a weighting to each analyst's estimate according to his or her past accuracy. SmartEstimate also gives greater emphasis to the most timely forecasts and less to those that have not been updated for a lengthy period.

Our screen lists the degree of change, up or down, in the SmartEstimate for each company, measured as a percentage of the original forecast. This column gives an idea of which way momentum is going for the stock.

The percentage difference between the SmartEstimate and the consensus estimate of the Street, called the mean, produces another StarMine metric called the Predicted Surprise ( ). When the two differ by a material amount, i.e. 2 per cent or more, the measurement gets the direction of a future surprise correct 70 per cent of the time, Mr. Gaumer says.

We began the list by screening for all gold companies traded in Canada with a market value of $50-million or greater, giving us an initial group of 118 companies. Applying the Predicted Surprise produced a list of just 24 companies that had positive results, several of them showing just a small fraction of a per cent.

The top three in this ranking were Anatolia Minerals Development Ltd. , Sulliden Gold Corp. Ltd. and Osisko Mining Corp. , with respective scores of 27.7, 15.0 and 10.7. But because none of the three has meaningful revenue yet, we disqualified them from our list. Several others were also removed from the list for the same reason: Andina Minerals Inc. , International Tower Hill Mines Ltd. and Crocodile Gold Corp. .

That left Winnipeg-based San Gold Corp. at the top of the screen, which also had one of the higher positive swings in its SmartEstimate over the last 30 days.

The last StarMine measurement applied to this screen is the Analyst Revisions Model, or ARM, which is a measure of the change in analyst sentiment ranging between 1 and 100. It looks at changes in the consensus over multiple time frames and not just for earnings, but also EBITDA and revenue revisions. It also takes into account the Predicted Surprise percentage shift on these various measures. When this score is near the top (100) or bottom (1) of its range, it is highly predictive of future earnings revisions and helps investors anticipate these events, Mr. Gaumer says.

Report Typo/Error

Follow us on Twitter: @GlobeInvestor


Next story




Most popular videos »

More from The Globe and Mail

Most popular