Dave Altig, Scott Baker, Jose Maria Barrero, Nick Bloom, Philip Bunn, Scarlet Chen, Steven J. Davis, Julia Leather, Brent Meyer, Emil Mihaylov, Paul Mizen, Nick Parker, Thomas Renault, Pawel Smietanka and Greg Thwaites.
The unprecedented scale and nature of the COVID-19 crisis has generated an extraordinary surge in economic uncertainty. In a recent paper we review what has happened to different indicators of uncertainty in the US and UK before and during the COVID-19 pandemic. Three results emerge. All of the indicators that we consider show huge jumps in uncertainty in reaction to the pandemic and its economic fallout. Most indicators reach their highest values on record, although the extent of the increases differ. The time paths also differ: implied stock market volatility rose rapidly from late February, peaked in mid-March, and fell back by late March as stock prices partly recovered. In contrast, broader measures peaked later.
Speculative buying can drive cryptocurrency prices down. This is contrary to the usual laws of economics. Blockchain technology limits how quickly transactions can be settled. This constraint creates competition for priority between different users. The more speculative activity there is, the longer it takes to make a payment. But the future value of cryptocurrency depends on its usefulness as a means of payment. Speculation therefore affects price formation through a channel that does not exist for other asset classes. This can explain the high price volatility of cryptocurrencies, and is consistent with the low adoption rate so far.
To consider Bitcoin volatility, we
look at 10-day returns (capital standards typically estimate market risk over a
10-day period) since 19 July 2010, when Bloomberg’s Bitcoin data start. We
compare Bitcoin with assets in three categories – currency pairs, commodities
and equities – and for each we have picked one low-volatility asset and one
more volatile asset. For currency pairs and commodities, we chose the most and
least volatile ones (in terms of standard deviation of 10-day returns) out of
the most liquid in each category. And we chose the most and least volatile FTSE
100 equities (again, in terms of standard deviation of 10-day returns).
For stable assets we expect a peaked distribution with short tails, as returns cluster near 0%. Figure 1 shows that Bitcoin has been more volatile than any other asset in our sample.
Figure 1
But people are often interested in the downside risk of assets. We therefore consider how Bitcoin’s Value at Risk (VaR) compares to other assets. VaR is the maximum loss over a given time interval under normal market conditions at a given confidence interval (eg 99%). A 10-day 99% VaR of -10% tells you that 99% of the time your 10-day return on the asset would be no worse than a 10% loss.
Figure 2 shows Bitcoin’s VaR is high, but the VaR of the other most liquid crypto-assets is higher. TRON’s VaR to date (-84%) is almost twice Bitcoin’s (-44%).
Figure 2
Giulio Malberti and Thom Adcock work in the Bank’s Banking Policy Division.
Comments will only appear once approved by a moderator, and are only published where a full name is supplied.Bank Underground is a blog for Bank of England staff to share views that challenge – or support – prevailing policy orthodoxies. The views expressed here are those of the authors, and are not necessarily those of the Bank of England, or its policy committees.
Often when analysing financial markets, we want to know the statistical distribution of some financial market prices, yields or returns. But the ‘true’ distribution is unknown and unknowable. So we estimate the distribution, based on what we’ve observed in the past. In financial markets, adding one data point can make a huge difference. Sharp moves in Italian bond yields in May 2018 are case in point – in this blog I show how a single day’s trading drastically alters the estimated distribution of returns. This is important to keep in mind when modelling financial market returns, e.g. for risk management purposes or financial stability monitoring.
Francis Breedon, Louisa Chen, Angelo Ranaldo and Nicholas Vause
Most academic studies find that algorithmic trading improves the quality of financial markets in normal times by boosting market liquidity (so larger trades can be executed more quickly at lower cost) and enhancing price efficiency (so market prices better reflect all value-relevant information). But what about in times of market stress? In a recent paper looking at the removal of the Swiss franc cap, we find that algorithmic trading provided less liquidity than usual, at worse prices, and that its contribution to efficient pricing dropped to near zero. Market quality benefits from a diversity of participants pursuing different trading strategies, but it seems this was undermined in this episode by commonalities in the way algorithms responded.
Volatility returned to markets in early February, sparked by strong US wage growth data. After months of calm, the S&P 500 equity index fell by 4% on 5 February and the VIX – a measure of US equity volatility that is sometimes referred to as Wall Street’s “fear gauge” – experienced its largest one-day move in its 28-year history. Interestingly, measures of volatility in other markets, including interest rates and currencies, moved by much less. So what caused the outsized spike in the VIX? Some of the rise was linked to rebalancing flows associated with VIX exchange-traded products (ETPs), which can amplify moves in the volatility market. The events have also led to some questions whether developments in VIX ETPs can also affect the S&P 500 itself –whether the ‘tail’ can wag the ‘dog’.
In 1995, Fischer Black, an economist whose ground-breaking work in financial theory helped revolutionise options trading, confidently stated that “the nominal short rate cannot be negative.” Twenty years later this assumption looks questionable: one quarter of world GDP now comes from countries with negative central bank policy rates. Practitioners have been forced to update their models accordingly, in many cases introducing greater complexity. But this shift is not just academic. Models allowing for a wider distribution of future rates require market participants to hedge against greater uncertainty. We argue that this hedging contributed to the volatility in global rates in early 2015, but that derivatives can also play an important role in facilitating monetary policy transmission at negative rates.
