The art of investing is such a deep mystery, even the most basic axioms don't really make any sense when you look closely.

Consider the axiom that you can only make money if you bet against the consensus. It means that if everyone agrees on how stocks should be valued, then whatever they agree on is already fully reflected in prices. So if the consensus is right, you won't make any money by going along with it. But if it's wrong, you could make a lot of money when the truth comes out.

Sounds good if you say it fast. But take it slowly.

Isn't the axiom that you can only make money if you bet against the consensus itself a consensus? Does that mean the only way to make money is bet with the consensus because the consensus says you should bet against the consensus? But how do you do that? Do you bet with the consensus by betting against the consensus because, after all, that's what the consensus says to do? How do you bet with the consensus and against it at the same time?

Even supposing you could unravel that paradox, how do you even know what the consensus is? Here's an example which, after all this philosophical meandering, is really the point of this column: gold.

I've written about it here most recently in late February. It's a long-standing market call for me, and I'm proud of it. In the world-wide market crisis of the past year, gold has been by far the best-performing risky asset.

But I'm told constantly that gold is a consensus trade, that it is a crowded long, which is to say that everyone knows about it and everybody is already in it. So if you are supposed to always bet against the consensus, you shouldn't be buying gold.

But is owning gold really the consensus view? How does one know that? Is there some kind of poll that is being taken?

I suppose price is itself a kind of poll. And since, as I said, gold has performed better than any other risky asset, then I suppose there must be some kind of consensus that it's a good idea to own it. But to me that doesn't really prove anything. Any asset that, say, tripled in value had to first double in value along the way. Should you have sold it when it had only doubled, because that was itself evidence that it was a consensus trade? Surely not.

Consensus is probably to a large extent a matter of anecdote. You can sense the consensus by what you read in the financial media, what you pick up at cocktail parties, and what you hear from fellow investors and traders. That means figuring out just what the consensus is can be a matter of subjective judgment.

I think that's especially true with gold. Unique among investment assets, gold is in some sense a cult. There are many investors who are simply fascinated by it, who make a kind of hobby of it, or perhaps even a religion. For these gold bugs there is never any reason to be bearish they find a reason why gold will go up no matter what is going on in the world.

With those guys always out there, there's a natural bias toward seeing gold as a consensus trade. But if those guys are indeed always out there, then it doesn't count. They can't really be part of a consensus if they can never actually make a decision to be for or against gold they're just always for it.

One client mentioned to me that gold must be a consensus trade because of those crazy infomercials urging people to send in their gold to have it melted down for cash. I don't see that at all. Those commercials wouldn't be on the air if they weren't successful in getting people to sell their gold. If there were a consensus to own gold, who would sell to some 800-number on TV?

And shouldn't consensus have something to do with the actual reasoning behind an investment conclusion? What if there is a consensus about gold, but for all the wrong reasons? Would it be smart to bet against such as consensus if you knew the right reason, even if it led to the same conclusion?

I think many people who invest in gold now are doing it because they think the world is about to end. They think the current global recession and credit crisis will lead to a collapse of government authority, to official currencies becoming worthless a world like the movie Mad Max. For them, gold is a hedge against disaster.

But I don't think disaster is coming. I'm betting on gold because I think disaster isn't coming. I think the central banks of the world, led by our own Federal Reserve, are going to save the world but at a price, the price of inflation. In my model, inflation is the one and only thing that determines the gold price. And inflation is coming.

But that's not a consensus view. Sure, I'm hardly alone in it. But it's not a consensus. Most economists would tell you that a world enmeshed in global recession is more likely to produce deflation, not inflation.

And they may be right. The talk Thursday coming out of the G-20 summit in London was that the International Monetary Fund would sell some of its vast gold holdings the IMF has the world's third-largest gold reserve is pure deflationism. Any time an investor dumps a tangible asset to raise cash in the face of emergency is a classic example of deflation.

But my bet is that the Fed and the other central banks won't stand by while potential IMF gold sales sop up even more of the scarce monetary liquidity in the world today. They'll print more money and drop it out of more helicopters and eventually overcome deflation with inflation. When that happens, gold soars to new highs.

To me, there's no consensus involved here at all. There's a very risky macroeconomic call to make. I'm calling it for inflation and I'm not alone in that, but there are plenty who are calling it for deflation. If I'm right, gold goes to $2,000. But believe me, I'm not taking any comfort from the idea that anyone else agrees with me on that.

For an alternate view on gold and why it's not a good bet against inflation, read Jack Hough's column .