top of page
Search

Analysis of the correlation of Bitcoin with the main Asset Classes


In 2022 all investment asset classes, apart from commodities for the first part of the year, lost value, and Bitcoin was no exception.

Many waved to point out that Bitcoin was not a deflationary asset, capable of protecting against inflation, because when inflation rose, its price fell. Other analysts argued that it correlated 1:1 to the Nasdaq, and therefore should be considered as a technology stock; many detractors have pointed out that it could not be considered a store of value, because its price has lost more than 70% (four times since its inception, I might add).


All true?

It depends on your point of view.


The Nobel Prize Albert Einstein demonstrated that in our universe everything is relative and depends on the point of view of the observer, and the developments of quantum mechanics teach us that the observer is part of the experiment and influences the results of the observations.

If we observe the trend of Bitcoin only in 2022, obviously we cannot argue that it protected investors against inflation, but if we consider that the so-called Crypto Winter, i.e. the period of cooling of the interest in bitcoin and digital assets in general, was widely expected for 2022 (as it will be for 2026), perhaps the observation period should be extended. If we extend the time horizon beyond four years, i.e. the duration of a bitcoin cycle, we note that the return has always been positive, and even well above the inflation rate.

Let us now pause to evaluate whether the statement that Bitcoin is an asset correlated with the Nasdaq and more generally with the financial markets is true.

Also in this case, if you want to have a statistically significant observation period, you cannot stop at the analysis of one year, but you need to broaden the spectrum of analyses.

The correlations can be calculated in two ways, either by analyzing the entire reference period available, for example 10 years, or by calculating the rolling correlation over 3-6 or even more months.

Analyzing the rolling correlation, for example at 6 months, means taking the returns of two different assets, from the period T0 to the period T183, and calculating the correlation, for that period. Subsequently, the correlation is calculated for the period T1 - T184, then T2 - T185 and so on for the entire observed period.

The result is a historical series over time that oscillates between -1 and +1 as in the graph below where the 3- and 6-month rolling correlation of Bitcoin against Gold has been estimated.


Fig. 1 Rolling correlation of Bitcoin and Gold


As can be easily seen from the graph, the correlation is far from constant and practically never above 0.5.

From reading this graph, can you argue that there is a correlation between Gold and Bitcoin?

Anyone who follows me, knows that I have always been critical of using correlation to optimize investment portfolios.

The reason is easily understood from this graph: does it make sense to use an indicator that varies with a higher volatility than the analyzed assets to make investment decisions?

To understand the dynamics in principle, as we are doing in this article, it can be a valuable indicator, but to optimize portfolios I would recommend you to avoid using it, because it can really lead to very dangerous "blunders".

Obviously the correlation changes a lot based on the frequency of the observation period, i.e. if I take daily or weekly data obviously the correlation is different and if I observe the value at 3 months or 6 months the trend will be different.

These considerations are very technical, but they are aimed at financial advisors, rather than their clients, because too often I still see professionals who optimize investment portfolios based on past returns, volatility (and so far steps) and past correlations.


Fig. 2 different calculation frequencies of the rolling correlations


As can be easily understood from this graph, the subjective selection of the parameters, with which the data is extracted, completely changes the final output and this is another reason why it is preferable not to use correlation data for portfolio optimization models.

But let's go back to the long-term analysis of the correlation between Bitcoin and the other main asset classes.


Fig. 3 Correlation matrix on a daily basis


While there is a very strong correlation between the American S&P500 index with the Nasdaq and the MSCI World, it is evident that in the long term the correlation of Bitcoin with respect to these financial markets is around 0.2, therefore very low.

This means, simplifying the concept, that if I insert an asset with 20% volatility in a portfolio with an overall volatility of 10%, and this asset is not perfectly correlated with the rest of the portfolio (correlation less than 1), the growth of the overall volatility will be lower than that which would result from the weighted average volatility of the assets in the portfolio.

This information is useful for understanding the weight to be included in an asset class such as Digital Assets which have a volatility 4-5 times higher than the stock market.

Finally and out of personal curiosity and I hope of the reader, we calculated the rolling volatility of Bitcoin against the same indices in the table above.


Fig. 4 - Rolling correlation between Bitcoin and S&P500


Fig. 5 - Rolling correlation between Bitcoin and Bonds


Fig. 6 - Rolling correlation between Bitcoin and Commodities


Fig. 7 - Rolling correlation between Bitcoin and MSCI World Equity


Fig. 8 Rolling correlation between Bitcoin and Nasdaq


As can be clearly seen from these graphs, the correlation of bitcoin with the Nasdaq was there, but slightly above 0.6, and it has dropped a lot in recent months; therefore, it certainly cannot be argued that the correlation between the two assets should be considered constantly high.

I also smile at some analysts who argue that Bitcoin is related to the monetary policies of the FED or to what Powell says, and even more those who argue that bitcoin has betrayed its image as a store of value, just because it fell in 2022 in a context of rising inflation. These analysts have forgotten that, following the cycles linked to the halving, Bitcoin is the asset that has grown by far the most in the last 10 years.

What should we say about the dollar or the euro which has been declining forever? I would first question their function as a store of value, especially in these years of high inflation!


Happy digital investments to you all

Daniele

21 views0 comments
bottom of page