Gold: A Modern Investment Framework For An Ancient Asset, Part III

In Part I of this series I derived a valuation framework for gold as a proven form of ‘insurance’ against potential losses in a portfolio of financial assets, and the currencies in which they are denominated. In Part II, I described a fair-value framework for modelling the price of gold. While the former approach implies that gold is significantly undervalued at present, the latter implies it is close to fair value. In this final part, I reconcile these two approaches by exploring the demand function for gold, and how it compares with that for money itself.

The Demand Function for Gold

We explored in Part II of this series how both long-term real interest rates and energy prices are key drivers for the price of gold. This is because they each represent an important aspect of the opportunity cost of holding gold as an idle, zero-yield asset, rather than investing in productive assets requiring capital and energy inputs. However, while relative cost (or valuation) is always an important factor when considering how much of something to own vis-à-vis something else, this should not be confused with the underlying, fundamental demand for that something. That is, when it comes to insurance, of course the premiums to be paid will be a factor when determining how much insurance protection to buy. But fundamental factors are also at play, in particular, perceptions of risk and risk preferences.

Such perceptions are always subjective and they are subject to change. How much fire, flood, theft, or other household insurance is demanded by customers will vary a great deal, regardless of price. Some customers will be proportionately more risk averse, willing to pay higher premiums for more protection, others lower premiums and less. Over time, some customers will become more, or less, risk averse, perhaps because of changing personal circumstances – perhaps for any number of reasons. This is the case regardless of cost, although cost will then be the final factor that clears the market.

When evaluating the fundamental demand for gold as a form of insurance, we consequently need to consider what underlying macroeconomic factors could shift risk preferences. This is quite a different exercise than merely looking at how the opportunity cost of gold moves up and down via real interest rates and energy prices.

This leads us in a difficult direction. As with official money, the underlying, fundamental demand function for gold is ultimately unknowable. What economists call ‘velocity’ is the inverse of the demand function for money. The lower the velocity, the higher the demand to hoard cash. Economists don’t like to admit that anything is unknowable, but they throw up their hands when asked to model velocity. Various attempts have been made to do so, but they always prove highly regime-dependent; that is, ultimately discontinuous and subject to qualitative factors. Trying to model gold’s demand suffers from the same problem.

Nobel Prize-winning economist Thomas Sargent explored extreme shifts in the demand function for money in a famous essay from the 1980s, “The End of Four Great Inflations”. But though fascinating, this paper is unsatisfying in its conclusions: the key takeaway is that measurable monetary and other economic variables, such as growth rates in monetary aggregates, cannot explain large changes in velocity such as those observed in large inflations. Rather, qualitative factors can, such as major shifts in government fiscal policies, or shifts in governments themselves, such as a change of party control or, more extremely, a political realignment or revolution. Sadly, as such factors are all so politically loaded and unquantifiable in any continuous or remotely objective way, any discussion thereof descends quickly into an ideological debate.

Sargent’s paper is useful, however, as it provides a means to reconcile the different conclusions reached in Parts I and II of this series. As I explained in Part II, the fair-value approach to modelling the price of gold, while potentially useful, has the notable limitation of being highly regime-dependent, and the regime shifts can be captured only in retrospect with binary-plug variables that are best understood as discontinuous, qualitative overlays on an otherwise continuous, quantitative model. Large shifts in the demand function for gold, as for that with money, are therefore regime-shifts, rather than movements along a given function. And as Sargent has showed, these regime shifts have to do with the fundamental nature of fiscal and monetary policies, and the credibility thereof.

Evaluating the Current Macroeconomic Context

Consequently, we need to consider the sustainability and credibility of the current set of monetary and fiscal policies. History is not exactly kind here. Debt levels (public and private) are unusually high. Financial system leverage is elevated, and regulatory capital requirements have recently led the Fed to re-implement a form of de-facto QE. The ECB is now buying corporate bonds, having run low on higher quality collateral. Japan appears to be slipping into recession notwithstanding an expansionary policy set. China’s economy is also slowing and was doing so before the coronavirus shock hit.

Given the above, it would be hard for a neutral observer to conclude that the general, perceived need for the insurance gold provides was at or below average at present. Rather, it would seem more likely that demand would be elevated. So why isn’t it? Well, given gold’s recent surge to over $1,600, the fair-value model actually does show a small overvaluation of around $100. But this amounts to noise – a variation of less than one standard deviation. And that is within a model that is itself regime-dependent.

In this highly unusual macroeconomic context, it seems curious that the price of gold is not significantly higher. But the same could be said of risk premia. Why are corporate spreads so tight? Equity valuations so high? Implied volatilities so low? I suspect the unsatisfying answer to all these questions is the same: markets are doing what markets do. That is, they are extracting risk premia where they can, following trends, and taking the policy cues provided by economic officials. As long as these are all pointing in one direction, gold is unlikely to outperform. But if they begin to point another way, for whatever reason, gold’s potential to re-rate is high.

To Conclude

Let’s explore one final consideration that one must take into account when purchasing insurance: timing – that is, to compare the value of a policy purchased prior to a damage or disaster event of some kind, and that of a policy purchased on the day after. Assuming it is honoured in full, the policy purchased prior to the event should be worth its full face value, less any premiums paid. As for the policy purchased after the event, well, its value is unknown, as it may never pay out. It might, indeed, be worth nothing. But, having suffered a loss uninsured, do you think the purchaser would forgo taking out insurance against the possibility of another such event in future?

In my view, this is yet another way to understand why gold has strong potential to outperform in the current macroeconomic context. Having been frightened by the global financial crisis of 2008-9, some investors may have subsequently acquired some gold as crisis insurance. Others may have thought about it, but didn’t follow through. But if another major financial crisis should occur again, for whatever reason, investors will notice a pattern, and the demand for crisis insurance as a core, strategic portfolio allocation, will rise accordingly.

John Butler has 25 years experience in international finance. He has served as a Managing Director for bulge-bracket investment banks on both sides of the Atlantic in research, strategy, asset allocation and product development roles, including at Deutsche Bank and Lehman Brothers.

(The commentary contained in the above article does not constitute an offer or a solicitation, or a recommendation to implement or liquidate an investment or to carry out any other transaction. It should not be used as a basis for any investment decision or other decision. Any investment decision should be based on appropriate professional advice specific to your needs.)