In just about any economic discussion you read or listen to these days, there’s one recurring theme- the strong consumer is picking up the slack.  Strong consumer sentiment and generational lows in the unemployment rate are just two of many examples.  A search of the term “strong consumer” on Google Trends also illustrates the strength of the consumer. While there’s still another nine days left in the month, searches for the term “strong consumer” in November are on pace to be the highest in at least a year.

So, we all agree that the consumer is strong.  Right?  Well, recently the market begs to differ.  The chart below shows the relative strength of the S&P 500 Consumer Discretionary sector versus the S&P 500 over the last year.  When the line is rising, it indicates that the Consumer Discretionary sector is outperforming the S&P 500.  However, when the line is falling it indicates that the Consumer Discretionary sector is underperforming, and underperforming is what the sector is doing now…in a big way.  Even as the S&P 500 is up around 4% in the last month, the Consumer Discretionary sector is down 1%.  While many traditional brick and mortar retailers that have fallen on hard times are in the sector because these stocks are already down so much, their weighting in the index has become very small. Meanwhile, stocks that have previously been big winners like Amazon.com (AMZN), Home Depot (HD), and McDonald’s (MCD) are the sector’s largest components.  Does the market know something we don’t? Sign up for Bespoke’s “2020” special and get our upcoming Bespoke Report 2020 Market Outlook and Investor Toolkit.

As the S&P 500 has pulled off of its record highs in the past week, investors have been more hesitant to label themselves as bulls.  This week’s survey of individual investor sentiment from AAII saw 34.24% of investors responding as bullish compared to 40.72% last week.  This was the biggest one week decline since an 8 percentage point drop in the first week of October.  After spending two weeks above it, this decline has also brought bullish sentiment back below its historical average. Additionally, this is the first time in two weeks that bullish sentiment was not the predominant sentiment level.  Despite this, the bull-bear spread is still in favor of bulls by 5.21 percentage points as has been the case for the past six weeks.

Although the bull-bear spread still leans positive, it has been narrowing over the past couple of weeks.  This is on account of bullish sentiment pulling back with those losses going to the bears. Bearish sentiment is now at 29.03%, up from 24.82% last week.  Bearish sentiment is still below (but now close) to its historical average of 30.36% as has been the case for the past five consecutive weeks.  This is the longest such streak since a seven-week run ending on May 9th.

As mentioned before, bullish sentiment no longer takes the crown for being the predominant sentiment. Instead, the highest share of investors, 36.72%, now consider themselves neutral.

Neutral sentiment has been above its historical average for 15 consecutive weeks now. This is just the eighth time in the history of AAII’s survey where there has been a streak of 15 weeks or more. Most of these have actually occurred in the current bull market with at least one occurring in each of the past five years except for 2018.  Even though the current run is long, previous streaks lasted much longer. The current streak is the longest streak since the one in 2017 that ultimately ended at 25 weeks long. Even that was not the longest streak on record though. That accolade belongs to the 41-week streak that lasted from 2015 through late 2016.  In other words, there is a historical precedent for extended streaks of above-average neutral sentiment.

Performance following past streaks once they reach 15 weeks has leaned on the weaker side with underperformance one month, three months, and six months later.  One and three months out the S&P 500 has actually averaged a decline.  On the other hand, the next week has typically experienced outperformance as has the next year, albeit to a lesser extent on average. Sign up for Bespoke’s “2020” special and get our upcoming Bespoke Report 2020 Market Outlook and Investor Toolkit.

Initial jobless claims last week came in surprisingly high at 225K.  That number has since been revised even higher to 227K.  Despite expectations of a decline to 218K, this week’s print was unchanged from the previous week’s revised number.  The past two weeks’ reports represent the highest level for jobless claims since June 21st’s reading of 229K.  Given the lack of improvement this week, the indicator remains above the past few months range, although the record streaks at or below 250K and 300K are still going strong at 111 and 246 weeks, respectively.

Given the increase in the seasonally adjusted number over the past two weeks, the moving average has also begun to tick higher, especially given the narrow range it has remained in recently. Rising to 221K, the four-week moving average is also now at its highest level since June’s when it reached 222.5K. This was also a slight increase, 0.25K, from the same week last year.

Turning to the non-seasonally adjusted data, jobless claims actually fell to 226.4K from 239K last week.  That is a bit of a break from the seasonal trend higher towards the year’s highs around this time of year.  This week’s reading was down year over year as NSA claims sit well below the average of 335.1K for the current week of the year since 2000.

Continuing claims were also a disappointment in labor data today as they rose to 1695K compared to estimates predicting no change at 1683K. As with initial claims, last week’s number was also revised higher to 1693K. Yet another parallel with initial jobless claims, continuing claims have been fairly flat over the past year.  As shown in the second chart below, the year-over-year change in continuing claims can no longer boast the same strength that it has since the start of the cycle. In the final week of September, continuing claims experienced their first year-over-year increase since 2010 when claims were still recovering from the last recession. Since then, claims have seen seven consecutive weeks rising versus last year. Start a two-week free trial to Bespoke Institutional to access our interactive economic indicators monitor and much more.

In addition to monthly statistics on residential construction released by the US Census earlier this week, there was also a quarterly update on attributes of newly started homes. The data tends to have pretty severe seasonality, so we have seasonally adjusted all of the series below.

Starting with who starts home construction, the vast majority of houses are built for sale by a builder. While there are some owner-built homes, contractors are about twice as common when it comes to single family houses. Multifamily starts are almost exclusively built for rental, with the number started for sale remaining near record lows.

Single family homes are almost always detached, free standing on all four sides. Homes that are attached to others but are still single family are less common.  Note the massive drop-off in detached during the housing crash.  We’re still not even back to average pre-housing bubble levels for detached single family starts at this point.

Among multifamily units, there used to be a respectable number of multifamily starts with less than 20 units, but over the last half decade or so virtually all starts of multifamily projects have involved projects with more than 20 units.

Finally, the Census also details the size of houses that have been started in the most recent quarter. As shown, from the 1990s to the mid-2010s, the median square footage of newly started homes rose by 20%. Since then, however, changes in taste and affordability have forced builders to economize: median square footage has dropped to the lowest levels since the 2010s and is still falling.  Based on square footage trends, it looks like “McMansion Hell” peaked in 2013 and slowly but surely people are coming to their senses.  More expensive land reducing footprints, less preference for space, and a desire to reduce sticker prices have all driven lower square footages.

For multifamily, the size of unit starts is a little bit more volatile and hasn’t tended to rise or fall consistently over time.  Sign up for Bespoke’s “2020” special and get our upcoming Bespoke Report 2020 Market Outlook and Investor Toolkit.