Gold (or the GLD ETF) edged a couple dollars lower last week. Gold mining shares once again overreacted to the downside (as they have on every sell-off in gold since last August) and edged lower by another 4%, according to the GDX gold mining ETF.

I don’t have a crystal ball, but the same signs that suggested a low was forming a week ago are still with us today. In fact, these “signs” are now getting even harder to ignore, even though prices haven’t moved all that much.

First, take the XAU/Gold ratio. One can question its relevance when trying to determine peaks and valleys in gold shares, but its usefulness as a historical indication of gold share valuations as either undervalued or overvalued relative to gold is unquestionable.

On Friday, the XAU/Gold ratio collapsed over 2% to a new 7-year low after Gold Fields (GFI) was slaughtered for 15%. This occurred after the company guided down production for the second half due to needed maintenance on two of its main mining shafts, which then triggered further deleveraging/liquidation in the rest of the large-cap golds as well. The move in the shares on Friday had nothing to do with the price of gold - but everything to do with selling begetting more selling in the wake of Gold Fields' collapse.

What that collapse has produced, however, is a gold mining universe that's probably more undervalued than it's ever been since the 20-year secular bear market in gold ended back in late 2000.

As you can see in the chart of the XAU/Gold ratio vs. the XAU (an index of senior gold mining stocks) above, the XAU/Gold ratio has never been this low during gold’s bull market, which began following the 2000 nadir. Valuations of the junior golds relative to gold are even cheaper than they were in 2000, believe it or not.

Some might say that this ratio has fallen because mining margins are contracting due to the high price of oil, which is a large component of mining costs. This was precisely the case from March through June, when the gold/oil ratio collapsed back to its 2005 all-time low.

Since June, however, this ratio has begun to climb (see the chart below of GLD, the HUI, the gold/oil ratio, and oil), meaning that gold mining margins are expanding once again. This is a trend I expect to continue, since oil will likely consolidate while gold moves to new all-time highs on renewed investment demand - which should occur during the seasonally positive period of the year for gold demand, which begins in August and lasts roughly through April.

Assuming gold continues to rally, what are the odds that this valuation disparity can continue with mining margins now expanding once again? I’d say close to zero.

Investors are being given a rare opportunity to buy mining stocks at these levels - but the window's quickly closing on that opportunity. As you can see from the XAU/Gold chart above, once a major low is put in for both the ratio and the XAU, the V-like move off the low in the mining shares is fairly swift, almost like a rubber band being snapped.

You might say: “OK, fine. Gold stocks are cheap. But what if gold falls? Why should gold rally, and why now?” In short, there are 4 primary reasons for gold to rally and rally soon:

1. Investment demand continues to increase. During July, which is typically a seasonally weak period for gold demand, fresh new investment demand took GLD’s gold holdings up to a new all-time of 706 tonnes, which made the GLD ETF the 8th largest holder of gold in the world behind the Japan’s central bank. On top of that, just this past week, legendary investor Jeremy Grantham said, "I bought my first gold last week, and I hate gold. It doesn't pay a dividend. I would only do it if I was desperate."

2. Real interest rates are negative all around the world, especially in the U.S., and will continue to be so as far as the eye can see. That's inflationary. It’s no coincidence that gold is rallying in all fiat currencies as well - not only in dollars, as some might believe.

Even now, gold is marching back to its recent all-time high in yen, euros, and dollars. As I’ve said before, this is another signal from the market that the fiat dollar-based monetary system that the world has enjoyed since 1980 is basically broken. And just as they did after the Bretton Woods system broke down in the early 1970s, investors will continue to flee to gold as a store of value until the new global currency system is identified.

3. With the U.S. banking system crippled, unemployment continuing to rise and the U.S. sinking ever further into the stagflationary soup, the Fed can’t even begin to think about raising interest rates to combat inflation. That means inflationary pressures will continue to build and support higher gold prices. I don’t care how high oil goes or how low the dollar goes, the Fed has never raised interest rates in its history while U.S. unemployment is rising.

And even if we didn’t know that, the Fed’s primary job is keep the financial system intact (as I've pointed out repeatedly since last August). That means as long as that system is at risk of complete collapse (as it is now), the Fed will continue to ignore rising inflationary pressures in favor of nursing the financial system with low interest rates and easy access to credit.

4. As for why gold should rally “now,” as opposed to in a couple of months or next year, the positive seasonal period for both gold and gold mining shares begins in August and lasts through next spring, a result of seasonal jewelry and Indian festival demand. When investment demand coincides with this seasonal demand, it creates a powerful upside move. Thus, the typical pattern following an early summer low (which in this case came in May) would now be for gold and gold mining shares to rise to new all-time highs sometime between now and the year's end, followed by further gains in the early spring of 2009.

Lastly, concerning timing: After writing bullish articles on gold and gold mining shares here on Minyanville over the past couple of years, I’ve noticed an interesting pattern regarding reader emails to my address. In particular, rare virtual “floods” of gold-related hate mail would seem to stream in near important lows - and then would suddenly stop. So I began to track the instances of mass hate-mailing back in 2007, which I've termed a “Hate Mail Buy Signal.” It's labeled in the chart below.

As you can see, I got two buy signals in a single week last week from this indicator - which is rare. The only other time this happened was in August of 2007, which proved to be a fantastic buying opportunity in both gold and gold mining stocks.

Will history repeat? No indicator is foolproof, but when taken in conjunction with all the other bullish factors currently in play, I tend to think it will.

For those that care, from both a valuation perspective and a risk/reward basis, I continue to like ASA (ASA), Newmont (NEM), Minefinders (MFN), Banro (BAA), New Gold (NGD), Gammon (GRS), Nevsun (NSU), Golden Star (GSS), and even Gold Fields, especially in light of Friday’s overreaction.

And contrary to what we saw in early 2008, I suspect mining shares are going to outperform gold on this next leg, due both to their current historic undervaluation and the rising gold/oil ratio, which will help mining margins to expand.