This article is an on-site version of our Unhedged newsletter. Sign up here to get the newsletter sent straight to your inbox every weekday
Good morning. Big rally on Tuesday! The bad-news-is-good news theory would hold that it was driven by belief that the situation as it relates to Evergrande is getting worse, so China will loosen policy. The good-news-is-good-news theory would say the key is that Omicron doesn’t look so bad. Take your pick and email us: firstname.lastname@example.org and email@example.com.
Is bitcoin just a high beta stock now?
Last weekend, crypto took a tumble. Bitcoin sunk 20 per cent to $42,000 before climbing back a tick. Ethereum and other coins followed.
In the wake of the decline, a popular narrative began to emerge: Wall Street’s big boys have arrived in crypto-land, and this drives stock-crypto correlations up. Bloomberg summed up the zeitgeist:
Wall Street pros and day traders alike are grappling with the ever-tightening links between crypto and mainstream markets: whether it’s the YOLO cohort trading in and out of meme stocks to whipsaw hedge funds, or institutional players driving bitcoin ups and downs to catch the retail crowd off guard.
From the FT:
“This started as a risk off move in traditional macro circles [which] triggered some liquidations in crypto,” said David Fauchier, a portfolio manager at digital asset specialist Nickel Digital, noting that investors who were selling equities also rushed to shed bitcoin which, unlike other markets, is tradable over weekends.
There is something to the idea that stocks and bitcoin are becoming more correlated, as you can see in this chart of 60-day rolling correlations between the S&P 500 and bitcoin (data courtesy of Coin Metrics):
Since Covid hit, bitcoin has more closely tracked equities. There has also been an uptick in the correlation in recent months. That has led some to declare bitcoin effectively a high-beta stock — a perfect toy for hedge funds.
But the positive correlation remains minor, as you can see on the chart’s y-axis. What’s more, typical correlation measures are designed for linear data. And bitcoin is anything but (have you seen its price graph?). Small correlations between non-linear data is thin evidence of a stable relationship. The rising association between bitcoin and equities could just mean they sometimes sell off together, but are otherwise unrelated. Or it could be pure noise.
The idea that Wall Street drove this past weekend’s sell-off is questionable, too. Because the crypto sell-off came after stocks dropped in the back half of last week, many saw contagion spreading via common ownership of stocks and crypto by Wall Street funds. But what seems more likely is the same jitters — Federal Reserve tightening and Omicron, mostly — scared two separate groups, crypto jockeys and proper fund managers.
We checked our intuitions with Philip Gradwell, chief economist at crypto analytics firm Chainalysis. He told us data from crypto exchanges didn’t support the idea that big Wall Street types were behind the weekend’s stumble. Unlike previous sell-offs, this time there were no abnormal inflows and only mildly higher outflows from large traders. The small role of big traders suggests the stock-to-crypto contagion theory is anecdotal at best.
As for crypto’s risk profile:
The big question people are thinking about is: is crypto a risk-on asset or a risk-off asset? Because it has fluctuated between the two . . . This weekend, it was a risk-on asset. Omicron is really driving fear and doubt in many markets. Crypto is not immune to that.
In May 2019, for example, bitcoin rose 15 per cent as stocks fell on trade-war fears. Now it is falling in line with stocks. An asset that moves both with and against risk attitudes is not really like a volatile high-beta stock. The closest comparison we can think of is penny stocks, trading on rumours and hype as well as real developments. But the likeliest case is that crypto is just a new thing. Pigeonholing it into one asset archetype or another is probably misleading. (Ethan Wu)
Can the Fed control long rates?
There has been a lot of discussion lately, here and elsewhere, about how inflation and growth are high, and the Fed is set to tighten, and yet long-bond yields are not going up much. Indeed the yield on the very long bond — the 30-year Treasury — is falling. This is sometimes held up as evidence that the Fed will make a mistake, tightening rates too late and as such violently, driving the economy into recession. Others think long rates will catch up eventually. I have, tentatively, argued that the market thinks the economy is on a sugar high, and will revert to normal quickly with just a mild nudge from the Fed.
But are inert long rates so unexpected? It’s not so clear. Start with the 30-year, charted against the Fed’s policy rate, over the past three decades:
The 30-year just goes down and to the right. It couldn’t care less what the Fed does, and never has. This will have been obvious to serious rate watchers. It’s consistent with most theories of how rates work, and it is widely acknowledged that policy has less effect on longer bonds than shorter ones.
Still, the state of current debate suggests the following point is worth making: let’s please just ignore the 30-year yield for the purposes of the current economic discussion. The 30 is gonna 30.
What about the rates that really matter, the 10-year and the 30-year mortgage rate? The two have broadly similar duration, despite their labels, because mortgage holders’ refinance option means it is a rare 30-year mortgage that lasts even 10 years. The chart:
This is more suggestive of Fed policy moving long rates around, because policy rate peaks correspond to peaks in both longer-term rates. But sustained examination raises doubts. In the past two cycles, the Fed increased rates for months before long rates responded at all (leaving at least one central banker famously puzzled). The Fed could be following the market, not leading it, or it could be some of each.
One noted sceptic about the Fed’s ability to control long rates, the economist Daniel Thornton, drilled down on the 2017-18 increase cycle in a note last year, looking at the policy rate and the 2- and 10-year bonds. His chart:
Thornton doesn’t even believe the Fed can control the two year:
One reason it is difficult to believe monetary policy caused the 144 basis point decline in the 2-year Treasury rate is the fact that 2-year Treasury rate had already declined 35bp by December 19, 2018, when the FOMC increased the funds rate target 25bp. Moreover, on January 4, 2019, chairman Powell said the Fed would be more patient about raising rates in 2019. It’s hard to believe market participants would take this as a signal the FOMC would start reducing the federal funds rate soon. In any event, the FOMC didn’t make the first rate cut until July 31, 2019. By then the 2-year Treasury rate had declined 109 of the 144bp.
Did the rates fall in anticipation of falling rates later? If so, Thornton points out, you would have expected the spread between the 10 and two to widen as both fell, as the two would be more sensitive to the change. Didn’t happen.
The moral of this story? Epistemic modesty. If there is a lot of surprise expressed today at the failure of long rates to respond to inflation and the Fed, there is also a lot of confidence in long rates finally increasing in 2022, under pressure from those two factors. But the relationship between long rates, policy and inflation is non-linear, controversial, poorly understood and generally tricky. Be careful out there.
One good read
Janan Ganesh elaborates on a theme we mentioned in this space on Monday — that part of the problem with politics today is lack of risk aversion, as the generation that remembers the abysmal mid-20th century dies off. He makes a financial connection, too, quoting Hyman Minsky: “Stability is destabilising.”