Even as the FOMC is still adding to its balance sheet, hawkish commentary from Powell and company has caused massive moves in the fixed income market resulting in a massive flattening of the yield curve.  Today alone, the spread between the yield on the 10-year and 2-year US Treasuries has narrowed by nearly 10 basis points (bps).  Even more extreme, is the fact that over the last three months the curve has flattened by more than 50 bps from well over 100 bps to just over 60 bps.

Looking at the three-month rate of change in the yield curve, it hasn’t flattened this fast in at least five years.  Last summer, we saw a similar but not as severe of a move.

Taking a longer-term look, flatter yield curves aren’t necessarily a negative economic signal, but there is an inverse relationship between the level of the yield curve and recession concerns i.e., the flatter the curve gets the more the market starts to price in the odds of a recession.  As shown in the chart below, every recession since the late 1970s has followed an inversion of the curve.

In terms of the pace of flattening in the 2s10s curve, the current pace has been the swiftest since January 2015, and longer-term, there haven’t been a whole lot of periods where the curve flattened at a faster rate.  Looking on the bright side, though, a rapid flattening of the yield curve by itself has not been a siren song for an impending recession.  As shown in the chart below, there have been numerous times over the last 40 years where the curve steepened by as much (red line) or more than it has in the last three months and the economy was nowhere near a recession. Click here to view and sign up for Bespoke’s premium membership options.

As equities continue to sell off, sentiment continues to tank in a broad sense.  That being said, there was actually a higher share of respondents to the AAII survey reporting bullish sentiment this week.  Coming off the lowest level since July 2020, bullish sentiment rose slightly and remains historically low at 23.1%.

Bearish sentiment, meanwhile, has surged 22.4 percentage points in the past month with more than half of respondents falling in the bearish camp this week.  At 52.9%, it was the highest reading since the week of April 11, 2013. In other words, sentiment this week among survey respondents was more negative than even the COVID Crash.  In all weeks since the start of the survey in 1987, there have only been 40 other weeks with as high if not higher of a reading.

Given the massive increase in bearish sentiment, the bull-bear spread has also outright collapsed. At -29.8, bears outnumber bulls by the widest margin again since April 2013.

Given bullish sentiment was actually higher as bears surged, the difference came from a big drop in neutral sentiment. Neutral sentiment fell by 8.4 percentage points to 23.9%.  That is only the lowest neutral sentiment reading since the first week of September of last year, but it was the biggest week-over-week decline since November 2020.

The AAII survey was far from being alone in indicating bearish sentiment.  The Investors Intelligence survey of newsletter writers saw the most bearish readings since the spring of 2020 and the NAAIM Exposure index also saw one of the more modest readings of the past year. Using all of these readings combined, we created our sentiment composite below to get a more general feel for sentiment across these indicators.  This week, that composite fell below -1 meaning on average sentiment indicators are a full standard deviation below (or in a bearish direction) their normal reading.   Of course, that was the first time such a reading has been observed since the COVID crash.

With data beginning in 2006, there have only been nine other times in which the sentiment composite has fallen below -1 for the first time in at least three months.  In the table below, we show those instances and how the S&P 500 has performed going forward. Overall, that bearish sentiment proves to be correct in the next week as the S&P 500 has declined two-thirds of the time. One month to one year later, though, returns have been biased to the positive side with positive returns at least two-thirds of the time.   Click here to view Bespoke’s premium membership options.

Initial jobless claims fell to 260K this week after last week’s higher than expected reading that was revised even higher to 286K this week.  Claims were expected to drop this week, but the actual decline was 5K larger than forecasts.

The beginning of the year typically sees a seasonal high in claims which abates into the early spring.  In fact, the third (last week) and fourth (this week) weeks of the year are some of the most consistent periods for week over week declines in the NSA number. Specifically, the third week of the year has never seen a WoW increase in the NSA number and the fourth week of the year (this week) has only seen claims rise 9% of the time; the fifth-best of any week of the year.

Given this, last week’s weaker than expected adjusted number was a result of a smaller than normal decline in NSA claims. This week’s drop of 73.3K to 267.6K was again smaller than normal, but much closer to the historical average of a 76.5K decline. In other words, the actual drop in claims before seasonal adjustment continues to be weaker than normal for this point of the year potentially as a result of high COVID case counts. Regardless, the actual level of claims for the current week of the year is basically right in line with the average from the few years prior to the pandemic.

Continuing claims are delayed an extra week to the initial claims number making the most recent print through the week of January 14th.  Claims rose that week from 1.624 million to 1.675 million. That was the third increase in a row, but all things considered, it was only a minor increase as the level of claims remains well below the range of the past several decades and is even below levels from the second half of December.   Click here to view Bespoke’s premium membership options.