Are crypto markets correlated with macroeconomic factors?

On a daily rolling three-month basis (chart 2) interest rates and the crypto index have  exhibited an inverse relationship 63% of the time since May 2017. This increases to 75%  from May 2020, following the start of the COVID-19 pandemic. The inverse relationship  is generally in line with how we would expect traditional assets to behave.

C. Is global money supply important for the crypto ecosystem? 

To assess the impact of global liquidity on crypto markets, we start by looking at M2 (a measure of money supply). M2 has surged since the Great Recession as central banks lowered interest rates and implemented QE. Chart 6 shows the relationship between M2 and S&P BDMI. M2 is the US Federal Reserve’s estimate of total money supply including all of the cash people have on hand plus money deposited in checking accounts, savings accounts, and other short-term saving vehicles. Since the 1980’s, M2 has exhibited exponential growth, particularly due to expansionary monetary policies during recessionary periods. In July 2022, M2 declined as the US reversed loose monetary policies.

Correlation between money supply and the crypto index is 0.75 over the historical period starting in 2017. This positive relationship notably broke down during 2018, as S&P BDMI fell, while M2 kept growing until July 2022. S&P BDMI’s decIine followed a crypto boom in 2017, when the market was flooded with initial coin offerings and new ventures, many of which ultimately failed. Both measures moved in the same direction in the second half of 2022. M2 contracted while the crypto market was hit by a series of events that fueled volatility and price declines, including the collapse of stablecoin UST in May and the downfall of cryptocurrency exchange FTX in November.

Chart 7 shows rolling three-month percent changes for M2 and S&P BDMI. Since 2017, changes for M2 have been generally positive, with a pronounced peak at the beginning of the pandemic, while the crypto index has exhibited several periods of negative returns. In the second half of 2022, both indices exhibited negative returns.

In general, crypto markets have performed well in periods of expansionary monetary policies, although we are not able to establish a causal relationship. Some of the large swings in crypto currencies have taken place following factors that are not directly related to monetary policy, such as the FTX collapse.

2. Does the perception of a possible incoming recession matter for crypto?

Growing recessionary risk could weigh on crypto assets if economic concerns dent appetite for higher-risk assets. At the same time, a recession perceived to be driven by poor government policies could arguably boost demand for crypto because the assets’ decentralized and borderless nature creates a potential shelter.

To understand whether recession risks weigh on crypto assets, we can use a widely followed gauge of economic expectations – the slope of the US Treasury yield curve. In particular, the difference between the yields on the 10-year Treasury Constant Maturity and the 3-month Treasury Constant Maturity, which has historically been a better signal of an incoming recession than other tenors. The yield on the 10-year Treasury is usually higher than that on the 3-month T-bill, giving the curve a positive slope. When this relationship inverts, it suggests that interest rates will fall in the future due to an economic slowdown. In the past, recessions have generally occurred whenever there was a sustained yield curve inversion. A notable exception is the current inversion that started in October 2022.

The yield curve has inverted three times since the 2008 recession. The first took place from March 2019 to October 2019, as above-target inflation prompted the Fed to increase interest rates, increasing expectations that the economy would fall into a recession. However, resiliency in the labor market prevented a recession during that period. The second inversion, at the beginning of the pandemic in February 2020, did not last long because of strong monetary and fiscal policies designed to mitigate the economic shock of COVID-19 restrictions. The yield curve inverted again in October 2022, as noted in the previous paragraph, and it remains in that state as of May 2023.

Chart 8 shows S&P BDMI and the difference between the 10-year and 3-month yields. When this difference is negative, the yield curve is inverted. The data is too short to find a consistent association between yield curve inversions and price declines for crypto.

Chart 9 below shows a more detailed view since late October 2022. Notably, S&P BDMI has more than recovered since losing 25% of its value in November 2022, even with the yield curve remaining negative. Some of the crypto volatility was linked to the implosion of digital-asset exchange FTX and its domino effect in the crypto lending space, as well as general uneasiness about crypto risks and governance.

The historical correlation between the daily returns of S&P BDMI and the inflation expectation indices is low, around 0.10.

Looking at rolling three-month returns for S&P BDMI and 10-year breakeven inflation expectations in chart 11 shows no conclusive pattern. There is a notable number of periods where returns on crypto and inflation indices exhibit opposite signs, meaning an increase in inflation expectations is not associated with an increase in cryptocurrency prices. However, there are also periods where the two measures are both positive or both negative.

Overall, the data to date does not support a conclusive answer on crypto assets’ hedging capabilities with respect to inflation. By contrast, Chart 12 below shows that gold prices and the 10-year Breakeven Inflation Expectation index have tracked each other quite well since 2003. Additionally, there is evidence of Granger causality between the 10-year Breakeven Inflation Expectation index and the S&P GSCI Gold index at a 95% confidence level. The same test fails for Bitcoin. (Granger causality is a statistical test to verify whether one variable is useful in forecasting another.)

In developing economies, there are cases in which crypto assets have been used as an alternative to holding the domestic currency amid very high levels of inflation, rapid currency depreciation, or stringent capital controls. This could underpin reports that emerging markets rank among the top countries for cryptocurrency trading (according to Chainanalysis data for instance) . Still, in some countries, cryptocurrencies may also be used to bypass financial sanctions, which complicates the study of crypto as a counter-inflationary asset.

It’s also worth noting that supply matters in crypto markets, even if this isn’t directly analogous to inflation. New coins are minted with proof-of-work mining or proof-ofstake validation. Supply of Bitcoin increases because miners get newly minted coins in return for their work. These rewards continually decrease and they will eventually end in about 2140. Ether, on the other hand, has recently had brief periods of burning more than minting. This will likely continue going forward.

4. What does a strong or weak dollar mean for crypto markets?

We measure dollar strength, using the Nominal Broad US Dollar Index, which tracks the currency against a weighted basket of currencies used by US trade partners. Investors should favor other currencies when the dollar is expected to weaken (an index decline) and demonstrate the opposite amid dollar strength (a rising index). Arguably, the same logic should apply to crypto assets, which would lead to a negative relationship between crypto prices and the US Dollar Index.

Chart 13 shows the Nominal Broad US Dollar Index and S&P BDMI. The historical correlation between their daily returns is –0.16. That compares with a –0.40 correlation between the US Dollar Index and S&P GSCI Gold. In chart 14, a rolling three-month correlation analysis shows an inverse relationship (negative correlation) between the US Dollar Index and both S&P BDMI and S&P GSCI Gold 75% of the time.

Idiosyncratic events may derail the expected relationship between crypto and the US Dollar Index over short time periods. Furthermore, because correlation does not substitute for causation, it is not obvious that a change in the US Dollar Index can provide insight into future movements in the crypto markets. In fact, there is no Granger causality between the US Dollar Index and Bitcoin prices.

5. Do financial stress and market volatility spill over into crypto markets?

Increased financial stress and market volatility are generally associated with declining crypto prices. We use the Financial Stress Index (FSI) to measure stress in global financial markets. The index incorporates five categories of indicators: credit, equity valuation, funding, safe assets and volatility. The index is positive when stress levels are above average, and negative when stress levels are below average. Chart 15 plots the FSI index and S&P BDMI.

FSI turned positive in early March 2020, around the day when WHO declared COVID-19 to be a global pandemic. The shock was felt in crypto markets, with Bitcoin losing more than 40% of its value that day. Many other cryptos also plunged, along with stocks. Bitcoin then subsequently started a bull run, along with other assets, which continued until late 2021. FSI turned negative in June 2020. The FSI turned positive towards the middle of 2022, as the aftermath of the Russia-Ukraine conflict increased prices of several commodities, putting upward pressure on inflation, and consequently raising interest rate hike expectations.

FSI turned positive most recently in mid-March 2023, though to a lesser extent than in 2022 or 2020. The failure of a few US banks added significant pressure to the banking system. These upheavals caused two major stablecoins (USDC and DAI) to depeg by 13%. Circle, the issuer of USDC, confirmed that $3.3 billion of cash reserves backing USDC were held at one of the failed banks. Both stablecoins recovered to their peg levels after the government confirmed that it would support the banks’ creditors. These events highlight contagion risks from traditional finance to decentralized financial systems. Interestingly, Bitcoin rallied after the event, leading to speculation as to whether this was a response to the banking crisis, to the government intervention or to something else.

We also looked at another widely used metric for market volatility – The Chicago Board Options Exchange’s (CBOE) Volatility Index (VIX). Also known as the market fear index, the VIX measures expectations of near-term volatility conveyed by S&P 500 index option prices.

Higher levels for the VIX indicate expectations for increased volatility. That adds pressure and fear to financial markets, drives liquidity premiums upward and reduces investors’ confidence. In such a market, it is expected that cryptocurrencies will exhibit more volatility.

In addition to VIX, we looked at the CBOE Crude Oil ETF Volatility Index (OVX), which measures the short-term volatility of crude oil as priced by the United States Oil Fund. We chose to analyze this index because considerable energy requirements mean that there may be a relationship between energy prices and Bitcoin miners. Chart 16 shows the VIX and OVX indices and S&P BDMI. The volatility indices both show sudden spikes in Spring 2020 during the COVID-19 period.

Crypto prices have an inverse relationship with the VIX and OVX indices – prices fall when the fear indices go up. S&P BDMI’s historical return correlation is –0.20 with VIX and –0.11 with OVX. Rolling three-month correlation is negative 90% of the time, reaching –0.60 in the past year. Interestingly, considering that both indices pertain to different asset classes, we found a Granger causality relationship between VIX and S&P BDMI.

The interconnectedness between traditional and decentralized finance is of particular importance because a healthy banking system is the backbone of a healthy economy. For example, in the US, the collapse of a few banks in March 2023 caused liquidity pressure in the banking industry. This in turn depegged some stablecoins, which are supposed to provide a stable bridge between the fiat and crypto ecosystems. There is no indication of how the wider crypto ecosystem would have reacted if the crisis hadn’t been contained. At the same time, crypto-friendly banks are exposed to crypto instability and have no control over market trends. That means the banking industry may be directly affected by uncertainties stemming from the crypto ecosystem.