Bespoke Baskets Update — July 2022
Best Performers Since Mid-June Low
Since June 16th, the market has reversed course higher, making the date at least a near-term bottom. From a technical perspective, the Russell 1000 broke through its 50-day moving average yesterday and broke above its upper downtrend line today. The 50-DMA is still moving lower, but this is the first time that the Russell 1000 has been above the 50-DMA since April 20th. However, the index is still 9.7% below its 200-DMA. These moves come as commodity prices have pulled back and earnings season begins.
As we highlighted in a Chart of the Day earlier this week, the lagging sectors during bear market declines tend to be the leaders in bear market rallies, which is holding true in the bounce since 6/16. The table below shows the 20 best performing stocks in the Russell 1000 Index since the 6/16 low. As you can see, these stocks are all still down considerably on a YTD basis, declining a median of 50.1%. However, these stocks have rebounded by a median of 42.1% since the low on 6/16. Only one of these stocks is below its 50-DMA, and only two are above their respective 200-DMA. If you think that the bear market has concluded, these would be some of the names worth looking deeper into.
On the other hand, the best performing sectors during bear market declines tend to be the worst performing sectors during bear market rallies. Of the 20 worst performing Russell 1000 stocks since 6/16, 14 belong to the energy sector (which has been the best sector on a YTD basis by a wide margin). On a median basis, these 20 stocks are still up 13.6% on a YTD basis, but they’ve shed 13.7% of their value since the market bottomed in mid June. Notably, these stocks (on a median basis) are closer to their 200-DMAs than they are to their 50-DMAs. If you think that we are currently in a bear market rally that is bound to reverse course, these names would be worth looking deeper into.
Declines Abound in Mortgage Data
Roughly one month ago, the national average for a 30 year fixed rate mortgage peaked above 6%. Since then, the rate has pulled back and stabilized around 5.75% in the past several weeks. Although mortgage rates have stabilized a bit, they remain at some of the highest levels since 2008 and have been heading marginally higher in the past week.
Given the slight rise in rates, demand for mortgages continues to predict further weakness in upcoming home sales data as we noted in today’s Morning Lineup. The MBA’s Mortgage Purchase index released this morning came in at a slightly lower reading than the June 3rd low for the weakest reading in the index since March and April of 2020. This week’s low is also below the range from the few years prior to the pandemic.
Not only has there been a lower volume of mortgages being applied for, but the actual dollar value of those loans has also plummeted amidst higher rates. As shown below, the average value of purchases reached a high of $460K back in March after a big increase in the first few months of this year. Since then, it has entirely reversed that move with the average value of loans having fallen all the way back down to $406K. That echoes the findings in other recent housing data which has similarly shown declines in home prices.
Higher rates have had an even more significant impact on refinance activity. Following another drop this week, the MBA’s index tracking refinancings has now reached the lowest level since November 2000. Click here to learn more about Bespoke’s premium stock market research service.
Chart of the Day – All You Wanted to Know About the S&P 500 Breaking Above the 50-DMA
Speculators Head for the Hills
As we do each Monday in our Closer report, we reviewed the latest data from Friday’s release of the CFTC’s Commitments of Traders (CoT) Report. This data shows how speculators are positioned in various assets based on positions in the various futures contracts. One asset that has seen some of the most notable changes of the report in recent weeks has been the S&P 500. As recently as one month ago, that data showed a net 1.29% of open interest was positioned long. Even though the S&P 500 has managed to come off the lows in the past month, positioning has collapsed with a massive move lower meaning there are far more speculators positioned short than long.
Whereas one month ago the reading was net long, in the latest data as of last Tuesday a net 9.3% of open interest was short. That made for the lowest reading since October 2015. In other words, speculators are positioned more pessimistically now than they were during the COVID crash, during the late 2018 near bear market, or any other time over the past several years. The change in positioning has also been rapid. As shown in the second chart below, that net positioning reading has fallen over 10 percentage points in just 4 weeks. Earlier this year in March there was an even larger 12.5 percentage point decline but prior to that, looking back through the history of this data, such large moves have been much rarer over the past decade than was the case in the 2000s and before. In fact, prior to this year the last time positioning in S&P 500 futures fell double digits in four weeks was all the way back in August 2011. On the other hand, the early 2000s frequently saw moves of this size if not larger.
Typically, sentiment data is considered contrarian in nature. In other words, pessimistic sentiment readings are followed by stronger forward performance of equities and vice versa. That has not exactly been the case for this CoT positioning data. Following past declines of at least 10 percentage points, returns have been mixed.
After declines of similar magnitude, the S&P 500 has experienced modest outperformance over the next week. One and three months out, however, have tended to be weaker with postive returns only half the time with negative returns on an average basis. Over the next month, each of the past four instances have been followed by declines.
Six and twelve months out have seen the S&P 500 higher two-thirds of the time, but average and median returns are not significantly better relative to all periods. Click here to learn more about Bespoke’s premium stock market research service.
Goldman and Citi Rip
Following earnings in recent days, both Goldman Sachs (GS) and Citigroup (C) have seen their stocks rip higher, resulting in rarely seen price moves for the stocks. Goldman’s trading segment benefited from increased trading volumes in FX, rate, and commodity markets, helping to fuel a beat on the top and bottom line. C also reported a top and bottom line beat, and every segment saw revenue rise y/y. Click here to view an in-depth analysis of these two earnings reports. Click here to learn more about Bespoke’s premium stock market research service.
Since the close on July 14th (three trading days ago), GS has ripped 12.5% higher which is the largest three-day rally since May 2020. On its earnings reaction day (7/18), the stock gapped up over four percent but finished the day with a gain of 2.6%. Today, GS jumped an additional 5.5% higher. Notably, the stock has now broken its downtrend and its 50-day moving average, a bullish technical sign for the stock. However, the stock is still down close to 20% on the year, which is worse than the performance of the broader financial sector ETF (XLF).
Following three-day gains of over 12% for GS (for the first time in at least a month), the stock has underperformed its all-period performance on a median basis across each time period we looked at (one day, one week, one month, and three months). Positivity rates are lower as well, so it’s hard to view this occurrence as an overly optimistic signal. One data point to take note of is that the average absolute three-month change is significantly elevated at 20.5% (median: 21.6%). For all periods since its IPO, the average absolute three month price change has been 12.9% (median: 10.1%). If the historical pattern holds, this would suggest that the next three months for the stock will be more volatile than normal. Click here to learn more about Bespoke’s premium stock market research service.
Citi’s chart looks quite similar, and its rally has been even more impressive than the rally in GS as the stock also broke above both the 50-DMA and the downtrend that has been in place since early January. With a gain of 17%, the stock has seen its largest three-day rally since the days coming off the COVID lows in March 2020. Both stocks have also struggled to definitively break through the 50-DMA, so investors should watch these levels carefully over the next few trading sessions. For both stocks, the 50-DMA is still moving lower, but the performance over the last few trading sessions certainly helps the technical picture.
In terms of forward returns, C has had mixed performance following prior three-day gains of at least 17.0%. The table below shows all prior occurrences since the October 1998 merger between Travelers and Citigroup. Of the thirteen prior occurrences, C’s median next-day performance has been a loss of 1.8%, but the stock has booked a median gain of 2.3% over the following week. Over the next three months, C underperforms all periods on an average basis but significantly outperformed on a median basis, and positivity rates have been narrowly above that of all periods. Three months out, C has averaged an absolute move of 24.5% (median: 11.8%), which like GS, is much higher than the 13.2% (median: 9.1%) average absolute three-month change for all periods. Long story short, although the technical picture has improved for both GS and C, investors should continue to expect volatility ahead.
Chart of the Day: When Laggards Lead
Bitcoin to Ethereum Ratio in Freefall
Risk assets have found some stability in the past month, but one asset that has been absolutely flying higher in Ethereum. The past week’s rally has risen to a remarkable 51% for the world’s second largest crypto bringing its market cap up from $126 billion to $187 billion. In that same span, Bitcoin has only gained 15.5%. After those moves, Bitcoin has returned to its 50-DMA while Ethereum has surged through the line.
Given that massive outperformance of Ethereum, the ratio of Bitcoin to Ethereum has collapsed. In fact, the ratio has now fallen for seven days in a row, tying three other streaks (September 2019, February 2020, and April of this year) for the longest streaks of consecutive days of declines.
The ratio has fallen considerably as well as consistently in the past week. Exactly one week ago, the ratio hit a near term high and has dropped 23.5% since then. As shown above, that brings the ratio down to the rough uptrend line that has been in place since late last year. Since the start of our price data for Ethereum in November 2017, there have only been a handful of other periods in which this ratio has fallen as sharply; meaning times when there has been similar underperformance of Bitcoin relative to Ethereum. The most recent of these was a little over a year ago in May 2021.
Below we show the performance of Bitcoin and Ethereum during each one of those one week declines of at least 20% in the ratio without another occurrence in the prior three months. This most recent drop in the ratio is only the sixth time such a move has occurred. Performance this time around has been middling relative to these other instances with Ethereum’s 43.9% rally through yesterday when the ratio crossed the 20% decline threshold roughly inline with the median performance of other instances (43.4%). Bitcoin on the other hand has actually been doing a little better with its 12.6% rally larger than the median of 9.8%.
As for where the two largest cryptos head from here, historical performance has been mixed. Overall Bitcoin has more consistently moved higher across the following year. One year out has perhaps been the period with the most dramatic outperformance of Bitcoin. Although it was only higher 60% of the time, Ethereum was only higher one year later once following the January 2021 instance. Click here to learn about Bespoke’s Weekly Crypto Report.
Chart of the Day – Volume Ain’t All It’s Cracked Up To Be
Homebuilder Sentiment Plummets
Another bad reading for housing hit the tape today as the NAHB’s Housing Market Index measuring the confidence of US homebuilders experienced its second largest decline on record behind the drop in April 2020. Homebuilder sentiment cratered 12 points month over month sending the index to the lowest level since May 2020 which is also one point below the late 2018 low.
Given the drop in the headline number, each of the individual components (present and future sales and sales traffic) similarly experienced historic declines month over month. Like the headline index, Present Sales and Traffic only experienced larger declines at the start of the pandemic. The same goes for Future Sales, although there was also a slightly larger drop in December 1987 making this month’s drop the third largest on record.
Geographically, no area of the country has been safe from plummeting sentiment, but the Northeast has held up remarkably well. Whereas the South and West saw double digit monthly declines that both were the second largest on record, the Northeast only fell five points to 57. That is a reading just shy of the top quartile of readings unlike the other regions that are in the 52rd percentile at best (South).
The lower sentiment for homebuilders is of course a function of rising mortgage rates and the subsequent dampening demand as a result. As for builder’s roles in existing homes, sentiment is also rolling over. In addition to today’s release, last week saw the quarterly release of the NAHB’s Remodeling Market Index measuring builder’s sentiment regarding remodeling projects. Unlike the Housing Market Index, this index remains historically elevated and down just a bit. In other words, remodeling demand has taken a hit but not to the same extent as projects to build a new home.
The most pronounced decline in remodeler sentiment has come from the projects carrying the highest costs ($50K or more) even though smaller project outlooks have also been falling for a few quarters now.
Backlogs have been unwinding as well even if they are historically elevated. A likely big reason for that has been a deceleration in new projects. The index for Appointments for Proposals is much less elevated in the 48th percentile compared to 56th for the headline reading or 68th for Backlog of Remodeling Jobs.
