From one of my longtime readers:

I just wanted to toss this suggestion your way and the motivation is partly selfish, but given the decline in gold the last 3-4 days (I actually exited all my long positions around 1500-1505 last Friday based on the breach of the technical support level at 1525-1535 and am now short in my trading account from that same level) I’d be interested to get your qualitative thoughts and maybe an update on your refined quantitative model with negative real interest rates and where it says gold should be trading.

If it turns out substantially above the current price of 1360, I’d be curious if you think that model isn’t valid or if gold is a bargain here. ?This article here got my wheels turning that bases on a gold price model on ratio to CPI:

http://www.marketwatch.com/story/golds-fair-value-is-800-an-ounce-2013-04-16?link=MW_story_popular

But to come up with an estimate of gold?s fair value, they calculate a ratio of gold to inflation going back as far as they were able to obtain data. They report that this ratio, when expressed in terms of the U.S. Consumer Price Index, has averaged about 3.2-to-1. Even at $1,400 an ounce, this ratio stands at 6.03-to-1, or nearly double this average.?

From a qualitative standpoint, the negative interest rate model made the most sense to me simply from a critical thinking standpoint. ?The relationship to CPI seems less reasonable to me if one starts with premise that gold is an alternative currency.

Anyways, thanks for any response or addressing this on your blog.

Links: The Gold Medal Gold Model, Gold does Nothing.

I updated my gold model.? This is what it looks like without re-estimating the parameters:

And this is what it looks like after re-estimating the parameters:

The real cost of carry in holding gold is negative, and it has been consistently negative for the last five years. and mostly so for the last ten years.? Thus the run-up in the price of gold over the last 10 years.

Now models are just that, models.? I can make three seemingly contradictory statements about this model:

• The old models did not predict the path of the gold prices well.
• The re-estimated models fit the data better than the old models.
• If the model is accurate, there is economic pressure to make the price of gold rise.

My hypothesis at this point in time is that easily tradable products based on gold encouraged speculative pressure, leading the price of gold to overshoot, and now it is correcting.? That said, when the real cost of carry is so negative, gold should appreciate.

Alternatively, we could try to develop a supply-driven model of gold, where we estimate the marginal costs of mining an additional ounce of gold.? Ore depletion is significant, but the effect is relatively constant compared to demand for gold.? It also helps to explain why the stocks of most gold miners have not done well, even with a rising gold price.

We often like to think that if a commodity price is rising, the stock of the producer must do even better.? Not always true, if the prices of extraction/production rise faster than the commodity price, as it has been with gold producers, the stocks will be a bad investment.

My final opinion is this: if you have a 5-year time horizon, I think you will do well with physical gold, where you take delivery, and store it yourself.? With easily tradable paper versions of gold, it is less clear, because you would need to analyze the actual assets.? There might be some credit risk involved.

I don’t think the currency devaluation competition is going away anytime soon, so gold will likely do well against paper.? The real question is when will some major country decide to give up and raise taxes dramatically, inflate, or default.? Aside from the raising taxes scenario, gold should do pretty well.? I might get less optimistic if the gold miners began making significant money,producing much more gold, but producing gold remains a hard business.