[Below is an excerpt from a TSI commentary published about two weeks ago. This discussion is being reproduced at the blog because it updates an opinion that was outlined at the blog way back in 2015-2016.]

Goldmoney (XAU.TO) originally was called BitGold and first began trading on the stock market in 2015. We wrote about the company four times at the TSI Blog during 2015-2016 (HERE, HERE, HERE, and HERE). The general theme of these writeups was: The company has a great product, but the stock is wildly overpriced.

Here’s how we summed up the Goldmoney business in the last of the above-linked blog posts:

“From the perspective of a Goldmoney user, the business is great. Customers can store gold, use gold as a medium of exchange and even take delivery of physical gold in manageable quantities, all at a low (or no) cost. From the perspective of a Goldmoney shareholder, however, the business is not so great. Of particular significance, unlike a mutual fund that charges a fee based on AUM (Assets Under Management), Goldmoney charges nothing to store its customers’ assets (gold bullion). This means that the larger the amount of Goldmoney’s AUM, the greater the net cost to the owners of the business (Goldmoney’s shareholders).

It’s important that under the current fee structure, Goldmoney will generally lose money on customers who use the service primarily for store-of-value purposes. This is where PayPal has a big advantage over Goldmoney. Nobody views their PayPal account as a long-term store of value. Instead, they view it as short-term parking for money to be spent, and when the money is spent PayPal usually gets a commission. This results in PayPal being very profitable, with earnings of US$1.2B (US$1.00/share) in 2015. Many of Goldmoney’s customers, however, view the service as a convenient way to store their physical gold. They don’t want to spend their gold, they want to save it.

Based on what I’ve seen to date I continue to believe that Goldmoney offers a great product, but is operating an inherently low-margin business deserving of a low valuation. Use the service, but don’t buy the stock.”

Since 2016 the company has grown a lot, mainly by acquiring similar or related businesses. Most importantly, it has modified its business model and now generates revenue/earnings from precious metals storage and lending. The fee structure is outlined HERE.

Over the same period the share price has trended down from highs of C$8.00 in 2015 and 2017 to a current level of C$2.18. Incredibly, the fundamental value of an XAU share is higher today with the stock trading near C$2 than it was in 2015-2017 when speculative fervour briefly caused the shares to trade as high as C$8.

Goldmoney Inc. now owns/operates two precious metals businesses called Goldmoney.com and Schiff Gold. Revenue for these businesses is earned as a weight of precious metal each time a client buys, sells, exchanges, takes delivery or stores precious metals through one of these businesses. Also, Goldmoney owns 37% of a jewellery manufacturer called Mene Inc. (MENE.V) and earns interest (in precious metals form) through the lending of precious metals to Mene. Lastly, Goldmoney owns/operates a company called Lend & Borrow Trust (LBT) that generates income by making fiat currency loans that are fully secured by precious metals.

The bulk of XAU’s earnings is in the form of precious metals that accumulate on the balance sheet. Furthermore, balance sheet assets not allocated to current working capital, investments and intangible assets are used to purchase and hold physical precious metals, the idea being that XAU’s holdings of gold, silver, platinum and palladium ounces will grow steadily over time.

With a Goldmoney account it is easy to buy and sell physical precious metals (PMs) at very competitive bid-ask spreads, with the PMs stored in secure vaults on an allocated basis (each client has ownership of specific pieces of metal). Also, it is possible to take delivery of your metal. Therefore, it could make sense to build up direct ownership of PMs via a Goldmoney.com account.

Alternatively, as long as the shares are purchased when they are trading near book value (BV), owning XAU shares is a reasonable way to build up indirect ownership of PMs. Owning the shares has the added advantage that if the company is well-managed then the amount of physical metal per share will increase over time.

The current BV is C$2.28/share including goodwill and C$1.79/share excluding goodwill. We think the latter number is the more relevant and therefore that the shares would be very attractive for long-term investment purposes at around C$1.80. However, the current premium to the C$1.79/share BV is not excessive, so if you are interested in XAU then it could make sense to take an initial position near the current market price of C$2.18.

[This post is an excerpt from a recent TSI commentary]

This year has been nothing if not interesting. Many unprecedented things have happened, one example being the performance of the US Industrial Production (IP) Index. As shown below, IP effectively fell off a cliff during March-April of this year. During May-July it climbed about half-way back up the cliff face before stopping in its tracks over the past two months. Nothing like this ever happened before.

When we say nothing like this ever happened before we are referring to the speed of the change. In magnitude terms the IP Index suffered a similar peak-to-trough decline during the Global Financial Crisis, but what took eighteen months during 2007-2009 took only two months in 2020. And after the 2007-2009 recession it took about two years for the IP Index to recover half of what it lost, as opposed to three months in 2020. In other words, what took 3.5 years during 2007-2011 took only five months in 2020.

The reason for the unprecedented speed of this year’s collapse is that the US economy didn’t fall off a cliff; it was pushed. The government (meaning: politicians and bureaucrats at the federal and state levels) deliberately crashed the economy. Policymakers then mounted such an extraordinary rescue attempt that personal income actually rose while the economy crashed and unemployment soared, which explains the unprecedented speed of the rebound.

The economic recovery stalled over the past two months (the IP Index for September was roughly the same as the IP Index for July), mainly because the government slowed the pace at which it was doling out ‘free’ money. The pace of the government’s money distribution is bound to ramp up after the November election, which probably will enable the economy to look strong during the first half of next year. However, there is no chance of a self-sustaining recovery. The main reason is that deluging the populace with newly created money does nothing to repair the damage caused by the lockdowns. On the contrary, it leads to capital consumption and sets the stage for another plunge into recession territory.

The biggest risk, however, isn’t that the ‘stimulus’ efforts won’t work and that the US economy will be back in recession within the next two years. That’s more of an inevitability than a risk. The risk of greatest concern is that policymakers will become even more aggressive in their misguided efforts to help and that these efforts will lead to hyperinflation.

For the first time since we started publishing these reports two decades ago, we cannot write that the probability of the US experiencing hyperinflation within the next two years is close to zero. The probability isn’t high, but it is significant.

[This blog post is an excerpt from a recent TSI commentary]

Moore’s Law, which is based on a comment by Intel co-founder Gordon Moore in the 1960s, states that the number of transistors on a chip doubles every two years. In effect, it states that the computer industry’s efficiency doubles every two years. Moore’s Law worked well in the semiconductor/computer industry for a few decades, but that was due to a set of circumstances that existed in that particular industry over a certain time rather than the general applicability of the ‘law’. The ‘law’ no longer works in the computer industry and can’t be applied in a useful way to technology in general. Wright’s Law, on the other hand, is more useful when it comes to explaining and predicting the effects of technology-driven improvements in efficiency. Wright’s Law pre-dates Moore’s Law by about 30 years (it was postulated by Theodore Wright in 1936) and states that for every cumulative doubling of units produced, costs will fall by a constant percentage.

The reason that Wright’s Law works better than Moore’s Law (Wright’s Law can be applied to all industries and has even been more accurate than Moore’s Law in the computer industry) is that it focuses on cost as a function of units produced rather than time. The beauty of Wright’s Law is that once an industry has been around for long enough to determine the relationship between the increase in units produced and the reduction in unit cost, accurate predictions can be made regarding what’s likely to happen over years and even decades into the future. The limitation is that a certain amount of history is required to establish the percentage reduction in cost that accompanies a certain increase in the production rate.

It should be possible to apply Wright’s Law to any growing industry. Of particular relevance to this discussion it should apply in the Electric Vehicle (EV) industry over the next several years, whereas it should no longer apply in the Internal Combustion Engine (ICE) vehicle industry. It probably won’t apply to ICE vehicle manufacturing in the future because the production rate of such vehicles appears to have peaked on a long-term basis. As evidence we cite the following chart of light vehicle sales in the US since the mid-1970s. This chart shows that the 9-month moving average of annualised light vehicle sales in the US peaked in 2005 and is lower today than it was in the mid-1980s.

Chart Source: advisorperspectives.com

Total light vehicle sales have peaked in many parts of the world, but EV sales are experiencing exponential growth. This implies that vehicle components that are specific to EVs, most notably batteries and electric motors, are going to get cheaper and cheaper. This will not only improve the economics of EVs in absolute terms, it will improve the economics of EVs relative to ICE vehicles.

Taking into account the life-of-vehicle cost, that is, the cost to buy plus the on-going costs to run and maintain, EVs already are competitive with ICE vehicles without the requirement for government subsidies, but within the next three years the economic benefits of choosing an EV over an ICE vehicle will become irresistible to most new car buyers in developed countries. This will happen with or without government incentives to buy EVs. It will happen because of Wright’s Law.

An implication is that current car and truck manufacturers that can’t figure out how to make EVs that consumers want to buy will disappear. Another implication will be a large increase in demand for the commodities that go into vehicle components that are specific to EVs. The commodities that spring to mind are the rare earth metals Neodymium and Praseodymium (NdPr), which are used in electric motors, and lithium, nickel and manganese, which are used in EV batteries.

Keep in mind, though, that the size of the EV market is limited by an overall light vehicle market that probably will be smaller in 15 years than it is today, because the combination of self-driving and ride-sharing will bring the age of personal car ownership to an end.

Wright’s Law naturally results in huge benefits for the average person, but central bankers think that they have to fight it. Even though prices naturally fall over time due to economic progress, central bankers believe that prices should rise, not fall. Therefore, they deliberately try to counteract the efficiency improvements that people in the marketplace are constantly trying to create. They do so by pumping new money into the economy or by encouraging commercial banks to lend new money into existence.

In general, the fast-growing industries that are focused on technological advancement are still able to reduce prices in the face of the central bank’s price-distorting efforts. This leads to the price rises being concentrated in industries where, due to government regulations or the nature of the industry, there is less scope for technology to drive prices downward. For example, Amazon.com was able to drive prices downward in the face of the Fed’s “inflationary” efforts, but most of its brick-and-mortar competitors were not. Other examples are the education and healthcare industries, where regulations and direct government ownership get in the way.

It’s hard to overstate the stupidity of a central bank strategy that is designed to make the economy less efficient. Currently we have the absurd situation in which the faster the rate of technological progress, the more that central banks do to create “inflation” and thus offset the benefits of this progress. They aren’t doing this because they are malicious, they are doing it because they are trapped within an ideological framework that prevents them from understanding the way the world works.

I discussed the relationship between gold and inflation expectations a couple of times in blog posts last year (HERE and HERE). Contrary to popular opinion, gold tends to perform relatively poorly when inflation expectations are rising and relatively well when inflation expectations are falling.

The relationship is illustrated by the chart displayed below. The chart shows that over the past seven years there has been a strong positive correlation between RINF, an ETF designed to move in the same direction as the expected CPI, and the commodity/gold ratio (the S&P Spot Commodity Index divided by the US$ gold price). In other words, the chart shows that a broad basket of commodities outperformed gold during periods when inflation expectations were rising and underperformed gold during periods when inflation expectations were falling.

This year, inflation expectations crashed during February-March in reaction to the draconian economic lockdowns imposed by governments and then recovered after central banks and governments tried to mitigate the lockdown-related devastation by showering the populace with money. This resulted in a crash in the commodity/gold ratio early in the year followed by a rebound in commodity prices relative to gold beginning in April.

The above chart shows that the rebound in the commodity/gold ratio from its April-2020 low was much weaker than the rebound in inflation expectations. This happened because there are forces in addition to inflation expectations that act on the commodity/gold ratio and some of these forces have continued to favour gold over commodities. Of particular relevance, the rise in inflation expectations has been less about optimism that another monetary-inflation-fuelled boom is being set in motion than about concerns that a) the official currency is being systematically destroyed and b) the private sector’s ability to produce has been curtailed on a semi-permanent basis.

Inflation expectations probably will continue to trend upward over the coming 12 months and this should lead to additional strength in industrial commodities relative to gold, but less strength than implied by the relationship depicted above. It probably will happen this way because more and more economic activity will be associated with government spending, which does nothing for long-term progress.

Eventually the relationship depicted above will be turned on its head due to plummeting confidence in both the government and the central bank, that is, at some point rising inflation expectations will start being associated with an increase in the perceived value of gold relative to commodities and pretty much everything else.