The Gold Bubble: End Game In Sight?

This week’s collapse in the price of gold was no more than I expected, and no less than I feared. I’ve never bought into the rationalisation of gold as an inflation hedge, or gold as a good investment.

My biggest worry with gold was that, as in the Genneva case last year, ordinary Malaysian investors were going to be caught in a bursting asset bubble, holding assets worth considerably less than they paid for. So far, that worry is slowly being realised – based on the current price, if you had bought gold anytime in Ringgit terms over the last two and a half years or so, you would’ve likely have lost money:

You’d still be in the black if you’d bought before that, but for how much longer is an open question.

The problem is that while gold does operate as an inflation hedge, the price is so volatile that for all intents and purposes it’s a pretty useless hedge, as this somewhat prophetic paper (it was issued in January) finds (abstract):

The Golden Dilemma
Claude B. Erb, Campbell R. Harvey

While gold objects have existed for thousands of years, gold’s role in diversified portfolios is not well understood. We critically examine popular stories such as ‘gold is an inflation hedge’. We show that gold may be an effective hedge if the investment horizon is measured in centuries. Over practical investment horizons, gold is an unreliable inflation hedge. We also explore valuation. The real price of gold is currently high compared to history. In the past, when the real price of gold was above average, subsequent real gold returns have been below average consistent with mean reversion. On the demand side, we focus on the official gold holdings of many countries. If prominent emerging markets increase their gold holdings to average per capita or per GDP holdings of developed countries, the real price of gold may rise even further from today’s elevated levels. In the end, investors face a golden dilemma: 1) embrace a view that ‘those who cannot remember the past are condemned to repeat it’ and the purchasing power of gold is likely to revert to its mean or 2) embrace a view that the emergence of new markets represent a structural change and ‘this time is different’.

As Carmen Reinhart and Kenneth Rogoff have argued, when people start saying “this time is different”, it’s time to head for the hills.

There are some fundamental factors supporting the price of gold, largely to do with its increasing financialisation i.e. the increasing use of gold as an investment vehicle. But that in turn means new sources of volatility, such as the use of leverage to purchase gold or gold-backed investment instruments. Demand for gold from China and India also underpins the market – according to one source, as much as 20% of gold demand comes from those two countries.

The problem is that even these structural changes don’t fully explain gold’s run-up in the past decade. The paper examines all the rationales for holding gold, from inflation and currency hedge, to gold role as a safe haven asset and for portfolio diversification.

The discussion in the paper focuses on two main criteria (the others are more or less quickly debunked) – inflation hedge and portfolio asset.

The data on gold as an inflation hedge is highly illuminating – gold appears to keep its purchasing power over the long term. Unfortunately, the long term here is defined in centuries, not years or decades, as the real (inflation-adjusted) price of gold tends to vary considerably over years and decades.

Also, in consequence, the implication is that gold’s very long run real rate of return is effectively zero, but rate of return can (and has been) both positive and negative over most realistic investment horizons.

That more or less jives with my own investigations into the relationship of gold with other commodities –their prices are very volatile in gold terms, exhibiting large multi-year price swings. I couldn’t find any stable long run relationship at all, but that’s probably because my sample period was too short (50 years compared to the 220 years the authors used).

Combine that insight with the current market price, and the tentative conclusion is that gold is sooner or later due for a very deep and very long bear market. The authors calculate a negative return of around -6% a year for 10 years (from the 2012 price) just for the real gold price to return to its historical mean – that translates to about US$780 in today’s prices. Of course, it could take much longer, or much, much shorter.

The second criteria – gold is underrepresented in investment portfolios – offers more hope to gold investors, though also offers grounds for greater concern.

Within the investment category for gold (the other two are industrial and jewellery), gold demand shows a remarkable tendency – it has a positive price elasticity from 2001-2011 i.e. an upward sloping demand curve. For the layman, what this means is that as prices increase so does demand, which is the exact opposite of the way demand should behave in theory. This offers some support for gold prices, in the sense that with the increasing financialisation of gold, more people have access to gold as an investment and desire to hold some, especially since it appears to have a very low correlation with other asset classes.

But, this also suggests a bubble mentality in the market for gold, which as in all bubbles, tends to end badly because the inverse is possibly true as well – as prices fall, demand falters, and prices spiral down further.

I’d tend to believe the latter is the more likely as the valuation of gold, as the authors remark, is very much like a Keynesian beauty contest:

The Keynesian beauty contest framework suggests that the price of gold is not determined by what you think gold is worth. What matters is, for example, what others think others think others think others think gold is worth.

Technical Notes:

Erb, Claude B. and Campbell R. Harvey, "The Golden Dilemma", NBER Working Paper No. 18706, January 2013