Interest in buying gold appears to be all the rage. As explained in Part I yesterday, the Google search request “How to buy gold” has reached a record.

To some, this helps to justify why the price of gold is near all-time highs of around $2,000/oz (£1,800/oz). But as I also explained in Part I, when you adjust for all the inflation through the years, the price of gold is lower today than it was decades ago.

At the time of writing, the war in Ukraine continues and may even be heating up. Energy security remains a key issue across much of the world. Central banks continue to raise rates into what appears already to be a simmering financial crisis. The US Federal Reserve has just come to the rescue of yet another major US financial institution, First Republic.

With interest in buying gold apparently so elevated, why is the price not even higher? At the end of Part I, I suggested this could be due, at least in part, to the structure of the international banking and financial system, including the regulatory regime.

Consider: much of the world’s wealth is held in official assets of some kind. Usually government bonds, most of the largest investment institutions are required to purchase them regardless of their price.

Let’s start with central banks. The US Federal Reserve purchases US Treasury securities in order to grow the dollar monetary base. The same is true for other central banks with respect to their currencies of issue.

While these purchases used to comprise only a small part of the outstanding stock of official assets, in recent years that has changed. The Fed today owns about 20% of the US federal debt. Other US government agencies hold another 25% or so.

Foreign official institutions – primarily central banks – also hold about a fifth of the US Treasury market. Thus over half of all US government bonds are held directly by governments or central banks.

The US is no exception. The Bank of England holds more than 40% of UK government bonds. The European Central Bank holds a similar amount of euro-area sovereign debt. The Bank of Japan holds over 50% of Japanese government bonds.

Bank regulations are “anti-gold”

Let’s now turn to commercial banks, which are required by laws and regulations to maintain capital and liquidity ratios in line with the international standards set by the Bank for International Settlements (BIS). Sometimes called the “central bank of central banks”, the BIS board is comprised of prominent national central bankers, as is much of its research staff.

While there is much arcane detail involved in calculating regulatory capital and liquidity ratios, there is one aspect that is simple to understand: government bonds are zero risk-weighted. Unlike private securities, such as corporate bonds or shares, the BIS rules allow for unlimited bank purchases of government bonds without any capital cushion whatsoever.

The reasoning is that, although government bonds can lose market value in the event that interest rates rise, they can’t default. Banks might take excessive interest rate risk and fail, as has happened to First Republic and several other US banks recently. But the BIS has nothing to say about that.

As it happens, gold is also zero risk-weighted, as gold cannot default. However, there is another BIS guideline of relevance here. Capital must be held against any asset that is volatile vis-à-vis the base currency. For a UK bank, this means that any non-sterling holdings must be risk-weighted; for a US bank, any non-dollar holdings, etc.

As it is not a base currency, banks must put aside a capital cushion against gold’s volatility. In other words, it is more costly for banks to purchase and hold gold than to purchase and hold their domestic government bonds.

Never mind that government bonds can be devalued by inflation and devaluation. In accounting terms, as they are denominated in their domestic banks’ base currency, they are zero risk-weighted, or “risk-free” as it were.

Gold may hold its real, effective purchasing power far better than national fiat currencies, but the BIS doesn’t take a view on that. In pure nominal accounting terms, sterling is sterling. Dollars are dollars. End of story.

Institutional “anti-gold” bias

Similar reasoning applies to the regulatory guidelines and governance of insurance companies and pension funds. Amongst the largest institutional investors on the planet, they are all required to match their asset holdings to their liabilities to a great extent.

Sure, they can and do take some risk around the margins, but in practice the bulk of their investment policies and processes strongly bias them towards matching the maturity of their assets – sometimes short, usually long – to liabilities.

They also have requirements regarding the risk management of their assets. It should be no surprise that their respective domestic government bond markets are seen as the lowest possible risk, whereas corporate assets seen at least at some risk of default are not.

Never mind that sterling, dollars, yen or euro government bonds might lose value in real, purchasing-power terms. Pension funds and insurance policies are written in a base currency. That currency might be devalued, but from a regulatory accounting perspective, the pension or insurance liability will be unchanged.

There is thus a strong bias towards domestic financial assets on the part of institutional investors. Gold might be a better long-term store of value than government bonds denominated in an inflating currency offering a negative real rate of return. But the regulations with which all must comply discriminate against gold.

In some countries there is a hard cap on how much gold investment institutions can hold. A few years back, a major Dutch pension fund was forced by its regulator to sell gold. The problem was that gold’s price had risen so far that its previously small holding had grown larger than the permissible limit. It appears the Dutch pension regulator will only tolerate gold holdings of less than 10%, notwithstanding some pressure from the industry to ease this constraint.

Let’s recap:

  • Central banks and government agencies hold huge amounts of their governments’ debts
  • Commercial banks needn’t hold any regulatory capital against possible losses on government debt
  • Large institutional investors are strongly biased for regulatory reasons towards holding domestic financial assets, including government debt
  • In some cases, large investors have strict limits on how much gold they can own.

Now, does that sound like a “level playing field” between financial assets and gold? Of course not.

“Anti-gold” bias suppresses both demand and price

Indeed, were gold not systematically discriminated against, investment demand for the metal would almost certainly be higher. Perhaps that investment demand would reflect the proven diversification benefits that gold has provided historically when included in a portfolio.

Let’s consider the reasoning of another Dutch pension fund that recently decided to acquire some gold and reduce holdings of government bonds. This change in investment policy followed on a historical study of prices by the fund which concluded, “This study showed adding gold to our asset mix had a clear added value because of its diversification benefits”.

If one runs the numbers using the past century of available data, what you find is that gold’s optimal weight in a portfolio of stocks and bonds is on the order of 10-15%. The precise amount depends on the specific historical starting point and base currency. But this order of magnitude holds for both sterling and dollars.

Intriguingly, many central banks hold more gold than they do foreign exchange reserves, including government bonds. Is there something they know that the regulators don’t? Or is that perhaps the entire point?

In any case, were gold to be more widely held, that would increase overall demand. But unlike national fiat currencies that can be electronically and physically printed into existence, gold’s supply is relatively fixed. It grows at around 1%/year, the limit of existing mining capacity.

Any positive shift in demand would thus increase gold’s price. Following the lead of the Dutch pension fund above, if more institutional investors decide that, as a matter of policy, they are going to increase their strategic allocation to gold, and pressure their regulators into allowing them to do so, then a much higher gold price would be the likely result.

Until next time,


John Butler
Investment Director, Fortune & Freedom

PS To learn more about the outlook for gold, Eoin Treacy will be holding a gold-focused webinar later this month – keep an eye out for more details on that soon.