MBS Live Day Ahead: Here's What a Bond Rally Costs in Terms of Stock Selling
Anything more than 20 points is a big day of selling in terms of the S&P, but that's a small enough number to be relatively common, periodically. From there, a sell-off of more than roughly 33 points becomes much less common. By the time we're talking about anything over 50 points, examples are limited to only a handful every year (and NONE from August 2011 through August 2015). Sell-offs of more than 80 points have occurred exactly 3 times during the recovery from the Great Recession--all of them in 2018--and yesterday was one of them.
The other two 80+ point S&P sell-offs were back in February, and we can basically throw the 2nd one out as being driven by whipsaw that was part of the same sell-off (aka, stocks bounced bigly on the way down and then returned for only a bit more selling). That leaves us with only one solid example for the purposes of comparing the current move.
When we looked at the longer-term chart yesterday, we noted that bonds didn't move much at all by comparison. In relative terms, that's very true, and the same was true yesterday. But back in 2018, the bond rally--while still not nearly commensurate with the stock sell-off--felt a bit bigger and better. Indeed it was! 10yr yields ended up moving nearly 25bps lower. So what gives?
Simply put, the 2018 stock sell-off occurred over 5-8 days, depending upon where you want to mark the start/stop. The biggest day of selling happened when stocks were already near their lowest levels in a month. The overall move was closer to 300 points whereas yesterday's move didn't quite make it to 150 peak to trough (including overnight futures levels). This begs the question: is our current rally about half the size of the February bond rally then? The short answer is "yes, mostly!"
But there's a catch, and it's not really a promising catch as far as past precedent is concerned. In terms of closing levels, the February bond rally was only about 10bps. That suggests intraday panic against the backdrop of a market that still hates bonds. Moreover, remember that 2nd day of more than 80 points of stock selling from February? Bonds didn't rally at all on that day. That means that the 25bp intraday move seen in the chart above had occurred in concert with only the first big day of stock selling.
Bottom line: additional stock selling from here wouldn't necessarily guarantee more bond buying, at least as far as past precedent is concerned.