Today's TDV Blog is written by friend and colleague, Robert Blumen, who is a repeat guest contributor to our blog. Robert is a software developer in San Francisco who writes for the target="_blank">Lew Rockwell.com, and other financial and economics web sites.

Why Value Investors Hate Gold

Jeremy Grantham, a highly respected and successful value investor target="_blank">Elsewhere, Grantham when discussing his recent purchase of a few ounces of the despised commodity, sarcastically said:

I hate gold. It does not pay a dividend, it has no value, and you can't work out what it should or shouldn't be worth...It is the last refuge of the desperate.

Warren Buffet, who needs no introduction, when asked by an eight grader whether gold should be part of a value portfolio,

James Grant publisher of , is less well known among the general public but has achieved cult-like status among professional investors. Grant does not hate gold; he might even be called a gold bug. He shows his customary wit in referring to the yellow metal as "the value investor's guilty pleasure." We like it but we shouldn't.

Why do value investors hate gold? To understand this, let's first look at how value investors think, and then understand why they hate gold.

Value investing is based on the premise that financial securities -- or really any kind of asset or business has two prices: the market price and a theoretical price arrived at through analytical methods known as intrinsic value. Value investors believe that the market should -- and probably will -- value an asset at its intrinsic value; when an asset trades at a much different price than its intrinsic value, the market is making a mistake. These errors are usually temporary in nature, giving the value investor an opportunity to capture a profit.

There are three ways to capture this differential: buying undervalued assets, short selling over-valued assets, or doing both at once. For the purpose of this article, we will focus on the long-only strategy; many of the most successful value investors over the years have used this method.

An entire discipline has grown around the analysis of intrinsic value, but to keep things simple all methods come down to either assigning a price to the cash flows produced by the security or breaking the security up into assets and liabilities that can be priced on external markets.

To understand valuation by cash flows, consider a bond that pays interest and principal payments. The calculation of intrinsic value relies on a best guess of the future payments emanating from the bond and applying the appropriate discount rate applied to derive present values from future cash flows. The discount rate could be the individual investor's own subjective rate of time preference. For example, a person who needs cash right now for some other reason might assign a very high discount rate to any investment opportunity. A more objective approach is to use interest rates or yields prevailing in the market (possibly averaged over a suitably long period of time) for similar assets to arrive at a comparable price for the cash flows of the asset being analyzed.

Breaking a company into smaller pieces that can be priced individually is the other major approach to valuation. A corporation has assets and liabilities. The assets may represent entire business units or individual properties that can be priced on an external market for similar businesses or properties. This is what the analysts for TDV do when they value a gold mining equity based on the value of its mineral deposits. This approach is an alternative to the cash-flow based approach for projects that do not yet have an income stream, but still have economic value because the assets have the potential to produce economically valuable outputs in the future.

The founder of value investing, Benjamin Graham, preferred to buy companies that traded at a market capitalization less than the value of their cash on the books. These companies were the safest, he thought, because there is little difficultly in valuing cash itself, and even less in arbitraging it. It does not require the investor to make too many assumptions, other than, that the management will not find a way to squander the cash.

But assets other than cash can be valued when there are external markets that provide a price. To take the gold mining industry as an example, a deposit can be valued based on the market valuations of the ounces or pounds of a defined resource (in the ground) for similar deposits to the properties being valued. A factory can be valued based on the land and the structures.

A business that generates steady revenues over a period of years is likely to do so in the future. The development of a mineral deposit into a mine is much more uncertain. But modern value investors have developed methods incorporate risk into the valuation model. Suppose, for example, one in five natural resource properties at a specific stage of development go on to be become mines. Then the market "should" price the property -- and therefore the stock -- at 1/5 of the value of a similar operating mine (discounted for development time). If the analyst shows that the property is trading at 1/10th its mine value, then the property is considered theoretically under-valued even though there is still a lot of risk involved. An investor would be taking a lot of risk by putting all of their capital into a 1/5 event, but a large enough portfolio of 5:1 odds purchased for ten cents on the dollar is likely to do quite well as the winners more than outperform the losers.

With an understanding of intrinsic value, it should be clear how the investor can profit from it. All valuation ultimately derives from cash flows returned to the investor. Even assets that are priced on external markets derive their value on those markets for their ability to generate cash flows in the present or in the future. If an asset truly has the ability to deliver the investor a present value of $10 per share, and if the investor is willing to hold it for long enough, they should eventually receive the $10 (it may be more than $10 by the time they get it depending on how long the wait). If the analysis is correct and the asset is really worth $10, then eventually the market "should" eventually figure this out and then reprice the asset to $10, short-circuiting the waiting time of the investor.

It is the ability of the asset to deliver the dollars to the investor that should make the price converge to intrinsic value. Given a correct analysis of intrinsic value, then in the worst case, the investor might have to wait for a while. A better outcome from the investor's viewpoint is that the market understands the earning power of the asset sooner rather than later, and the market price reflects this. Some corporations try to force the market to revalue the asset by issuing or raising a dividend or by spinning off subsidiaries into stand-alone firms. Each firm then must have its own price while before they may have been lumped together.

Here is the reason that Grantham and his ilk are so disdainful of the yellow stuff: they cannot value it the way that they value stocks and bonds. Grantham and the like hate to buy anything they can't value because they might be over-paying for it. Buying something without a quantification of its worth falls into the category of irrational speculation, not much different than tossing a coin. In their terminology, it has no intrinsic value. For an asset to have intrinsic value, it must have something that can be priced on a market external to itself (either cash flows or a balance sheet); gold has neither.

But if gold has no "intrinsic value" does that mean it has no economic value? Should the wise investor only purchases assets that can be valued by Graham's methods? Does that make the gold buyer a crazy speculator?

I have observed that some value investors identify value investing with economic rationality itself: any purchase that is not backed up by an intrinsic value is not only baseless speculation but an act of pure irrationality. In my view, their mistake is that value investing is not reality, it is a model; it works, like any model, subject to certain assumptions; one must take care to apply the model within its boundaries. Intrinsic value is a

The difference between fundamental analysis of the commodity and a security is that none of the data can provide a number for the future price; however rationality extends to things that cannot always be quantified. But analysis of supply and demand can provide the investor with some rational reasons for expecting the future direction of the price to be higher or lower, which is enough to make an investment decision. Is it totally irrational to think, for example, that a commodity that has no new mines coming on line, declining supply of existing mines, and increasingly popular uses will rise in price? If the people who built mines were strict value investors who insisted on an intrinsic value for every decision, then no mines would ever get built.

I have been discussing the valuation of commodities in general by supply and demand fundamentals, but when it comes to gold (as I wrote in target="_blank">current function as a shadow money that competes with global fiat currencies for remonetization should the fiat money system shoot itself in the head. The more suicidal the fiat money system becomes, the greater the demand to hold gold as a store of value should fiat money fail in this function.

Investors can make a rational decision about gold. Clearly, analysis of the gold price must be based on a forecast of demand to hold the metal itself, competition from demand to hold other assets, and demand to hold fiat money. Some of the metrics that fit this approach (and have been used by others) are: the growth over time in the quantity of gold compared to the quantity of fiat money; the DOW index-to-gold ounce ratio over time, and the total value of gold portfolio holdings compared to the capitalization of other financial assets globally. Typically these measures showed historical extremes in gold's favor in the late 90s/early 2000s and have moved back closer to historical averages.

But is there any reason to think that gold should revert to historical means, or even overshoot them, in the same way that we think that an asset "should" trade at its intrinsic value? What keeps the value pricing model in line with market pricing is the economic arbitrage: the investor can always hold the asset and eventually receive the intrinsic value, if they wait long enough.

Is there an arbitrage that will bring gold back in line with past measures of purchasing power? Or are we just looking at historical trends that may or may not recur? My answer to this has two parts.

Measuring gold against the money supply or the DOW is a measure of purchasing power. To the extent that gold is valued as shadow money, we can expect its purchasing power to converge on a value approximating its historical purchasing power as money, adjusted for the growth in the size of the world's economy relative to the quantity of gold.

But there is a more profound and less easily quantifiable argument for holding gold. Earlier I mentioned the finite limits in the domain of applicability of value investing. Value investing is an example of what the economist Ludwig von Mises called Chief Editor