Originally published on October 18, 2018
One of the main narratives around Bitcoin is the Digital Gold or Store of Value (“SOV”) use case. Naturally compared to Gold, which has historically proven to be the best store of value. With institutions starting to move in (Yale, Fidelity etc), another narrative emerges around Bitcoin as an institutional investment asset. These two narratives often go together, but it makes sense to explore them separately.
TL;DR
- By looking at both Gold and Bitcoin, we need to separate the SOV narrative from the institutional investment case
- Store of value is meant to preserve purchasing power and hedge against inflation, not to generate wealth per se
- Being a good addition to an institutional portfolio is determined both by the asset’s correlation with the existing assets (the lower, the better) and its expected returns (the higher, the better)
- Both Gold and Bitcoin fit the bill, but the potential opportunity in the latter has a much higher risk-reward
Let’s explore.
GOLD AS INVESTMENT
Why do institutions use Gold in an investment portfolio? Two reasons: 1/ good returns and 2/ low or negative correlation with other asset classes.
While stocks have had the best returns historically, Gold’s performance has been inline with US 10-year treasury bonds and lower than other types of bonds like corporate or high-yield. (data source: Deutsche Bank). Gold has beaten the more general commodity complex, which tends to be highly cyclical.
The chart below shows returns in real terms, i.e. adjusted for inflation.
Portfolio hedge. Gold has low and often negative correlation vs. major asset classes over a long time period.
Gold’s rolling correlations vs S&P 500 and 10-year US bond yields (proxy for bonds).
Why is this important? Diversified investment funds need to generate returns on a risk-adjusted basis. One important metric which is often used is the Sharpe Ratio. In simple terms, it shows the excess portfolio return above the risk free rate, divided by the portfolio volatility measured as the standard deviation of the returns. The higher the ratio, the better as the portfolio is generating more return per unit of risk. If you add to a portfolio an asset with low correlation to the other assets in that portfolio, it decreases the volatility of the total portfolio and the Sharpe Ratio increases. This is often valid even if the asset’s returns are lower than the existing assets. A higher Sharpe Ratio is good news for funds.
As an example, see the table below with the full article here.
Even though Gold itself has a much lower Sharpe Ratio, adding it to the portfolio increases the overall ratio.
Further, Gold is not the only inflation hedge available to fund managers. Stocks and real estate and land and other commodities are also inflation hedges. Investors buy Gold for many reasons, among which its lower correlation, being an inflation hedge, industry and jewelry demand etc.
The pure inflation protection use case is crucial if you keep fiat cash in the bank, which leads us to…
GOLD AS STORE OF VALUE
For something to store value, it has to protect its holder from inflation and to preserve his or her purchasing power. In this sense, Gold is compared to currencies and not to productive cash-flow generating assets like stocks. Gold has done very well in contrast to fiat currencies over long time periods.
Gold is perceived as a store of value because: a) of its physical properties and b) of its ability to preserve purchasing power leading to its “psychological” property to be universally accepted as valuable money for millennia.
Interestingly, Gold has often been as or more volatile than stocks and EUR/USD (see chart below). On some occasions, the difference between Gold and fiat was up to 6x. For reference, Bitcoin is currently ca. 10x more volatile than fiat, but during the period from mid-2015 till end-2016 it was as volatile as Gold during its 70s/80s peaks. This is showing that a perceived store of value can be volatile.
Moving on to…
BITCOIN AS STORE OF VALUE
A good case as to why Bitcoin is better than Gold has been laid out in this essay by
In a nutshell, if we as society agree that gold has value (even if we can’t eat it or use it to buy groceries), then we can apply the same logic to Bitcoin. In addition, BTC is more divisible, can’t be confiscated (at least until the government has memory extraction devices), is easily transported, can be universally accepted electronically, can be transferred using a cell phone and we won’t even need the internet soon (mesh networks, satellites) etc.
For Bitcoin to preserve purchasing power vis-a-vis fiat money, it has to increase in price over time. Since its inception, this has obviously been the case, but irrelevant this early in the game because: 1/ the speculation angle has been dominant and 2/ the store of value use case has been more widely accepted over the past few years only. The effect in developed countries is less prominent because inflation runs is only a few %. Anecdotal evidence from countries with hyperinflation like Venezuela suggest that people are ready to switch to Bitcoin (or others) and abandon fiat.
BTC has been much more volatile than Gold since its inception, but as its acceptance increases, the volatility will decrease. It is interesting that the pattern looks similar, but it may simply be a coincidence.
We are in the first years of the “BTC as a Store of Value” hypothesis and yet to see how it will play out. Gold has a few thousand years head-start and has proven itself. Every year in which Bitcoin increases in value in real terms (adjusted for inflation) against a base fiat currency, one’s value has been successfully stored.
BITCOIN AS INVESTMENT
How big is the SOV use case? $7trn Gold market. Extend to include some forms of fiat and we get $15trn (
’s work). Go even wider up to $100trn (post by
). Given its current market cap of ca. $135bn, clearly there is some way to go. The reality is that nobody knows, but the fact that we can logically come up with such potential market means that the risk-reward is viewed very favorably. This is why many people suggest keeping a small % of one’s net worth in Bitcoin — just in case. For any early holder of BTC, such a development would indeed be akin to a gold mine. Obviously, once (and if) BTC settles as a SOV, its role and return expectations would change a lot. But until then…
The low correlation of BTC to other major asset classes coupled with its much higher return expectations (compared to Gold and frankly most, if not all other assets even after its already astronomical rise) makes it a much desired part of any investment portfolio (check the last paragraph in the Yale article).
Let’s look at some charts.
Below you can see the rolling 1-yr correlation of gold and BTC with the S&P 500 index. BTC has had an even narrower correlation range than gold.
Similar low correlations with European and Emerging Markets stocks.
Low correlation to other major currencies.
Low correlation with depreciating currencies of countries in crisis — Turkey and Argentina for example.
Low correlation with gold itself, but much much higher return expectation.
Low correlation with bond yields.
A fair question is what would happen with these correlations if/when institutional investors and central banks fill their portfolios with Bitcoin. Depending on how they classify it, it can become more correlated with other assets. When one type of investor base becomes dominant, then investment flows, redemptions, portfolio allocation, structured products and other factors start to play a much larger role. Another fair question is how this wall of money will want to apply control and influence over the asset. It will be fascinating to see how it plays out.
In addition to direct investments, there are many similar ways to invest around Bitcoin and Gold.
Unlike Gold, BTC is programmable money and adding new functionality over time will enable more ways to explore, develop and invest — this is a crucial differentiation.
Finally, a head-to-head performance chart, just for fun. Gold has been flat since BTC started to gain mindshare, even though this is very likely to be a mere correlation and not causation 🙂
Thank you for reading