Great Investment Articles

A repository of great articles to help make informed investment decisions.

Thursday, March 15, 2007


John Hussman: Gold/XAU Ratio Signals Buy for Mining Stocks

Note: The Gold/XAU ratio closed over 5.0 earlier this week, and the headings are ours, not Hussman's.

Part 1 -- Why the S&P 500 isn't a great investment for the next decade

Near-term considerations for the stock market aside, I am sometimes asked what the prospects are for stock returns perhaps 5 years ahead. In general, a 10-year horizon is more reliable because speculative influences wash out to a greater extent, but some observations on this may be useful.

I've noted for some time that S&P 500 earnings are at the very top of their long-term 6% peak-to-peak growth trendline – a level of earnings that has typically been associated with an average price/earnings multiple of 10 (not the current 17). See last week's market comment for a review of these conditions. Meanwhile, the dividend yield on the S&P 500 is about 1.9%.

We can imagine a few reasonably optimistic outcomes. First, suppose that record profit margins are persistently maintained, so that earnings continue to grow along the very peak of their 6% growth trend. Rather than assuming the price/earnings multiple on these peak, record-margin earnings will fall to anywhere near 10, let's assume that 5 years from now, the multiple merely touches 15 (just two points lower than presently). Given that assumption, the 5-year S&P 500 total return would work out to be:

1.06(15/17)^(1/5) + .019(17/15+1)/2 – 1 = 5.41% annually.

Alternatively, we could assume that given extremely wide profit margins and rising unit labor costs, earnings will be flat over the next 5 years. We can identify many historical periods where earnings did not, in fact, grow over a 5-year period. Indeed, for earnings to be flat over the coming 5 years, assuming continued revenue growth at 6% annually, profit margins would still have to be above their historical norms 5 years from now. But let's not be too dour. Let's also assume that the price/earnings ratio on the S&P 500 will increase to 19, which would still be a rich valuation on earnings at that point. Given those assumptions, the 5-year S&P 500 total return would be approximately:

(19/17)^(1/5) + .019(17/19+1)/2 – 1 = 4.05% annually.

In my view, neither of these assume particularly negative outcomes, but they imply quite unsatisfactory long term returns, which underscores the point that rich valuations rarely deliver pleasant long-term results.

In order for the S&P 500 to achieve a “normal” annual return of about 11% over the coming 5 years, we have to assume a maintenance of record margins, sustained top-of-channel earnings growth (which has never before been sustained for such a period), and an expansion of valuations to a multiple of 20 times peak earnings (the same multiple as at the 1929 and 1987 peaks, which is double the average historical multiple on top-of-channel earnings). Investors should think now about whether these assumptions are plausible, because they may find themselves wondering later why they ever did.

My impression is that the probable expectation for total returns on the S&P 500 over the coming 5-years is below 5% annually, in a likely interval that includes zero. It takes implausibly optimistic assumptions to move substantially above that range.

As for 10-year returns, for which the historical evidence has typically allowed tighter confidence intervals, the following chart updates the study that appeared in the February 22, 2005 market comment (“The Likely Range of Market Returns in the Coming Decade”) using the same methodology. Note that actual market returns moved outside of the typical range only during the late 1990's bubble, and that the most recent 10-year return of about 7.6% since 1997 has been at the top of the expected range precisely because current valuations are at the top of historical norms.

Currently, the likely range for S&P 500 returns over the coming decade is between a -3% annual loss and a 5% annual return, centering in the low single digits. That range will seem preposterous to some investors, but remember that it took the late 1990's market bubble to move actual returns even 5% outside of this set of bands. Unless investors anticipate a repeated excursion into similar valuation extremes, it would be a good idea for them to recognize now, rather than later, that stocks are unlikely to produce satisfactory long-term returns from current valuations.

Part 2 -- Why mining stocks are a great investment now

For now, both stocks and bonds appear priced to deliver relatively unsatisfactory long-term returns for the risks involved. Stock market investors are still expressing a preference to speculate, at least on the basis of price/volume behavior, but we are not far from the point where stocks could again be characterized as overextended.

In the Strategic Total Return Fund, we continue to hold a short-duration investment stance, mostly in Treasury Inflation Protected Securities. The Fund also holds about 20% of assets in precious metals shares. It's worth noting that the fairly simple but generally useful Gold/XAU ratio is now pushing close to 5.0, though it has not breached that level.

To reiterate my remarks on the Gold/XAU ratio from the May 2, 2005 comment:

“To put some historical context on this measure, since 1974, the Gold/XAU ratio has been greater than 5.0 about 15% of the time. When the ratio has been this high, the XAU has followed with annualized gains of 89.6%, on average – a figure that remains high even if the data is split into multiple samples. When the ratio has been greater than 4.0, the XAU has followed with average annualized gains of 27.4% (though the finer profile of returns has been sensitive to other conditions such as interest rates, economic trends, and inflation). In contrast, when the ratio has been less than 3.0 (meaning that the gold stocks are very elevated relative to the actual metal), the XAU has declined at an annualized rate of -36.6%, on average.

“Importantly, the return/risk profile for precious metals shares is strengthened further if the economy is experiencing weakness. For example, when the Gold/XAU ratio has been greater than 5.0 and the ISM Purchasing Managers Index has been less than 50 (indicating a contracting U.S. manufacturing sector), gold shares have appreciated at an average annualized rate of 125.6%. In contrast, when the Gold/XAU ratio has been less than 3.0 and the Purchasing Managers Index has been greater than 50, precious metals shares have plunged at an average annualized rate of -49.9%.”

Such strong periods for gold are also generally associated with weakness in the U.S. dollar. Something to think about as the economic picture evolves in the months ahead.


May 2006   June 2006   July 2006   August 2006   September 2006   October 2006   November 2006   January 2007   February 2007   March 2007   May 2007   July 2007   October 2007   February 2008   March 2008   October 2008   October 2009  

Great Investment Articles Home    Email GreatInvestmentArticles

Great Investments Blog     Great Trades Blog

Disclaimer: Great Investment Articles may have a position in all or some of the stocks discussed in this blog, but is not paid by any company to promote their stock. Great Investment Articles contains opinions, none of which constitute a recommendation that any particular security, transaction, or investment strategy is suitable for any specific person. Great Investment Articles does not provide personalized investment advice.

Enter your email address in the box below to get emailed any new blog entries (within an hour or so of an update). Your email address won't be listed or sold.


This page is powered by Blogger. Isn't yours?