Can Interest Rate-driven Crypto Selloffs Be Hedged?

It has been a very tough stretch for cryptos. In the past week (as of 2:50pm ET on 7-Jan-22),

  • Bitcoin – 11.4%
  • Ethereum -14%
  • Solana -19.1%
  • Avalanche -16.4%
  • Polygon -17.6%

And on and on. But cryptos have only mirrored some of the high-valuation, high-growth, momentum-oriented stocks. Week-to-date, some returns include:

  • iShares S&P 500 Growth ETF (IVW) -4.1%
  • ARK Innovation ETF (ARKK) -10.4%
  • Global X Cloud Momentum ETF (CLOU) -8.4%

But the reason is pretty simple. All of those – cryptos, momentum stocks, profitless tech – are just very much levered to rates. The Fed signaled this week that it was going to tighten policy faster than many expected, Treasury yields have spiked, and a lot of the high-fliers have been seriously wounded.

So what is a crypto investor to do

Most people buy cryptos because they think they’re going to go higher, not necessarily as a hedge to a portfolio or another asset class. In fact, one of the big selling points is that crypto is an uncorrelated asset class. But as I argued in December, the opposite is true – cryptos are just super high-beta assets. They just swing (quite widely) along with the markets over time.

But let’s pretend that crypto investors were interested in limiting risk. Without jumping into complex derivatives to limit up- and downside exposure, something like a pair trade between long crypto + short Treasuries would somewhat lock in the crypto-only exposure.

Does that work in practice though?

To preface: This is just a sample of one hedge, and over an unusual period for Bitcoin at that…

I looked at the price of Bitcoin and the price of the iShares 20+ Year Treasury ETF (TLT). If Bitcoin’s price rises, you gain. If long-duration Treasury yields rise (that is, Treasuries sell off), TLT loses. But if you’re short TLT and yields go up, you gain (there are other ETFs that are short Treasuries, but we’ll pretend we’re borrowing for free for this example). A long Bitcoin/short Treasury strategy should work because they should be natural hedges to each other.

If we look at a snapshot over the past week, the decline in Bitcoin’s price would have been offset somewhat by the short TLT position, as yields have jumped. Of course, I’d never expect TLT to fully offset Bitcoin’s ups and downs since cryptos are inherently much more volatile than Treasuries. But maybe it would help create a more stable portfolio.

Unfortunately, this one example didn’t work

From 6-Jan-20 to 5-Jan-21, $100 in Bitcoin alone would have ended up at $560. A $100 investment in TLT would have led to $106.50 over the same period – so a short TLT position would have netted you a $6.50 loss (we’re ignoring borrowing costs for simplicity).

Let’s say you invested in a 130/30 long Bitcoin/short TLT portfolio. That is, you borrow 30% of your position in TLT to roll into Bitcoin. That would have give you $697 at the end of the period. You’d also have increased your Sharpe Ratio – that is, your risk adjusted return – from 1.81 with Bitcoin alone to 1.82 for the long/short portfolio. However, you can’t get very much more levered without eating into your returns. In fact, a 110/10 portfolio would have give you a Sharpe Ratio of 1.81 – better than the 130/30 and about the maximum risk-adjusted return possible. A 200/100 portfolio would give you Sharpe Ratio of 1.15 – not great.

The problem is, you can’t really get the volatility of the mixed portfolio down by combining with a short TLT position. The Bitcoin portfolio had an annualized volatility of 0.75 (yikes!). The TLT was 0.182. The 130/30 portfolio is 0.90. That is – adding a short position did nothing to reduce your portfolio’s volatility. You have to have a portfolio that is long both Bitcoin and TLT to reduce your portfolio’s volatility – but even then, you’re going to give up some of your risk-adjusted reward.

This is probably all a function of a very unique period in Bitcoin’s history – and it may just work yet

If Bitcoin settles down and somehow becomes less volatile, the long/short strategy might just work to earn a significantly better risk-adjusted return. But Bitcoin futures are up only 5% over the past year to $41,665, and in that time there has been a closing high of $67,985 on 9-Nov and a closing low of $29,815 on 20-Jul.

However, since mid-2021, we’ve seen a broad decline in the volatility of Bitcoin – meaning it might just become tame enough to play some long/short positions against it.

Here’s a link to the spreadsheet if you want to play around with the numbers – you can change the weights of the portfolio by adjusting O14 on the calc tab (highlighted in yellow)