February 03, 2025

Digital assets – evolution and correlations with other asset classes

View the report
Subscribe to our Research
  1. Authors
  2. Overview
  3. Research and Insights
Robin Marshall

Robin Marshall, MA, M.Phil
Director, Global Investment Research
linkedin
Alex Nae

Alex Nae
Quantitative Research Analyst, Global Investment Research
linkedin

Key takeaways:

  • We find that the Digital Assets Bitcoin and Ethereum show variable correlation of return coefficients with traditional assets, but these increased sharply with risk-on assets, like equities, after the Covid and inflation shocks
  • In addition, Digital Assets have shown high betas and volatility in response to macro shocks, and not yet developed stable store-of-value characteristics, like gold
  • Caution is required in interpreting the results, given the short time series on which they are based. Correlations may change as the blockchain settles down and transitions to a more stable regime, and as investors become more comfortable with investment use cases and notions of fair, or equilibrium value
  • This may contribute to lower volatility in Digital Assets over time, as the asset class matures
  • On the basis of the correlations found, Digital Assets cannot be classified as either commodities or currencies, and may constitute a brand-new asset class
  • Incorporating Bitcoin and Ethereum into a portfolio represents an opportunity for higher returns, but must be traded off against increased volatility, consistent with the principles of financial theory

Points of differentiation:

  • The paper draws on FTSE Russell data across multiple asset classes to assess the correlation of returns with Digital Assets
  • By extending the analysis to include selected Digital Assets, it complements our previous, 2024 paper on the correlation of multi-asset returns
  • It also enables us to assess how far Digital Assets share key characteristics with traditional assets, i.e. as a store of value, or safe haven, or as risk-on/risk-off assets

What does our research mean for investors?

  • The research enables investors to better inform their investment decisions about Digital Assets, drawing on the empirical evidence on correlations of returns from FTSE Russell / LSEG data
  • By identifying how correlations of Digital Assets with other asset classes have evolved, during the macro-economic shocks of recent years, it helps asset allocators assess how Digital Assets may respond to future shocks
  • Finally, the research suggests the tendency of financial assets to increase correlation, or co-movements, during negative market shocks, like Covid or the inflation shock that followed, may increase the appeal of Digital Assets for investors, given they show different performance characteristics

  • Digital Assets (DAs) are any assets that can be created and stored digitally, containing or storing value, and enabling digital transfer of ownership.

  • New types of Digital Assets include non-fungible tokens (NFTs), stable coins, central bank digital currencies, and security tokens for underlying financial assets.

  • Digital Assets now perform various functions on a blockchain, driving the growth of a heterogeneous asset class where different DAs serve different purposes.

  • Digital Assets encompass a broader set of assets, while cryptocurrencies are a smaller subset primarily used as mediums of exchange.

  • Key questions concern the relationships between cryptocurrencies and other asset classes, especially as cryptocurrencies have developed as investment assets in the global financial system.

    • On October 15, 2021, the U.S. Securities and Exchange Commission (SEC) approved the first Bitcoin futures exchange-traded fund (ETF)
    • On January 10, 2024, the SEC granted approval for 11 spot Bitcoin ETFs
    • On May 23, 2024, the SEC approved the first Ether ETFs

  • We examine the correlation of returns between DAs, and financial assets, since their inception, using FTSE Russell index multi-asset class data. In that regard, the paper should be seen as complementary to our earlier paper on the Correlation of Multi-Asset returns. Specifically, we examine:

    1. Cross-Correlations: The static and time-varying relationships between cryptocurrency returns and traditional asset classes, including equity indices (Russell 1000 and Russell 2000), fixed income asset classes, and industry components from these indices across technology, energy, and alternative energy.
    2. Rolling Correlations and Betas: The stability and evolution of these relationships over time is assessed, measured through rolling correlations and beta metrics, to capture shifts in market dynamics and potential regime changes.
    3. Macroeconomic Context: We also assess the broader macroeconomic context and its influence on the correlations between cryptocurrencies and traditional financial assets, with a focus on inflation, monetary policy and financial conditions.

  • FTSE Russell’s Digital Assets index solutions span the investable market from large cap to micro-cap, with flagship market cap, sector, or single asset indices. The modular  FTSE Global Digital Asset Index Series has been structured with investor flexibility and customisation capabilities in mind.

    For more information on our Digital Asset indices click here.

Related Research and Insights

Research

Digital assets – evolution and correlations with other asset classes

In this paper, we examine the correlation of returns between Digital Assets, financial and real assets, using FTSE Russell index multi-asset class data since their inception.

Research

The forgotten stimulus - uncovering the revival of Europe's peripheral markets

The EU’s Recovery and Resilience Facility is providing a huge fiscal boost to the Euro area and underpinning the recovery in its financial markets, but the impact diverges, with the size of RRF allocation and financial market composition being significant factors.

Research

Valuation Matters – are shifts in index composition impacting market valuations?

How have market valuations been impacted by changes in index composition? Can controlling for these changes improve the predictive power of valuations in forecasting forward-looking returns?