Seeking Alpha

Buy-And-Hold Vs. Rotational Strategy: Tracking Sentiment Strategy Results (November 2019)

|
Includes: ACN, ADP, BF.B, CSX, CTAS, EL, EXPE, FICO, GIB, HD, HSY, INTU, ITW, LOW, MA, MCD, MDT, MKC, NKE, NOC, NSC, PAYX, RMD, ROL, RSP, SBUX, SHW, SPLV, SYK, TFX, TGT, TXN, UNP, WMT, XMLV
by: Cory Cramer
Cory Cramer
Value, portfolio strategy, contrarian, medium-term horizon
Summary

Many investors find it difficult to sell high-quality winners.

But if the price gets high enough relative to expected future returns, even the stocks of the best businesses should be sold.

I've written 'sell' articles on 32 high-quality large-caps so far this year.

This article tracks the performance of those stocks versus some more defensive alternatives.

My goal is to demonstrate the long-term validity of my strategy of taking profits in overvalued stocks by understanding sentiment cycles.

Source

Introduction

This article is a follow-up to a series of articles I recently wrote about how to avoid losses and profit from sentiment cycles. Part one of the series "Ignore Sentiment Cycles At Your Own Risk" explained what sentiment cycles are, and how even the stocks of high-quality companies can become overvalued enough to sell. I also shared a working theory of the factors that I think contribute to the formation of a sentiment cycle with any particular stock. In part two of the series, I shared a long-only investment strategy that can help investors avoid some of the losses associated with a sentiment cycle by rotating out of the overvalued stock and into a more defensive position, then, when the price of the overvalued stock comes down, rotating back into the stock and being able to own more shares than when you sold it without spending any extra money.

For example, let's say one owns the stock of company XYZ and it trades at $100. The business is a great business, but the price has become so expensive that the implied future returns if someone bought the stock at that price are so low that it makes sense to sell it. Now, let's say there is a defensive ETF like the Invesco S&P 500 Low Volatility ETF (SPLV) which also trades at $100, but is likely to trade with much less volatility than the market, and unlikely to fall as far and as fast as an overpriced stock.

Let's say someone sells XYZ and rotates into SPLV while both are priced at $100. Then, over the course of the next several months, the price of XYZ comes down to earth and falls to $80, while SPLV stays at $100. If one owned 100 shares of XYZ initially, they can now sell their SPLV shares and buy 125 shares of XYZ because the price is cheaper. This results in a 25% 'free share gain' compared to if one had just held the stock of their great business throughout this entire period.

That's a basic explanation of how free share gains and a long-only rotational strategy work. Back in 2018, I wrote a long-running series about how to do this with highly cyclical stocks, and I continue to update that series each quarter. This new series is going to be about the stocks of businesses whose earnings are not highly cyclical. These businesses will all have earnings with low-to-medium earnings cyclicality, but they will be stocks that have become overpriced mostly due to the sentiment changes of the market.

Simple Rotational Approach

In the last explanatory article of this series "When To Sell And When To Buy Back Again," I explained several different levels of sophistication an investor can take while using a long-only rotational strategy. If you would like more details on the strategy, give that article a read.

For this series, I am sharing what I call the "simple mixed rotational strategy." The main goal, other than warning investors that their high-quality stocks will probably not produce great returns if they are held at high prices, is to demonstrate the usefulness and effectiveness of the rotational strategy in real-time rather than using a backtesting approach. Backtests can be useful, but not nearly as useful as watching how a strategy works in real-time.

Now let's outline how the strategy works.

The first step is identifying a high-quality business with a great long-term history of consistent and steady earnings growth. All of the stocks I write about in this series are stocks that I am at least interested in owning for the next ten years. These are not low-quality short-selling candidates. Occasionally, I eventually find something I don't like about the business and hold off buying it even after the price falls, but initially, all the stocks in this series appeared high enough quality to interest a potential purchase from me. I don't own any of the stocks in this series personally, yet, but my primary audience when I share the articles are investors who already own the stocks. My goal is to let them know if their stock is overvalued enough to sell, with the ultimate goal of buying it back at a lower price and increasing the number of shares they previously owned for free.

The second step is to identify if the stock is expensive enough to sell. In April of 2019, I started specifically examining stocks that looked overvalued on the surface to see if they were sell-worthy based on their expected 10-year forward returns. I call these articles "10-year, Full-Cycle Analyses." About 2/3rds of the stocks that I examined did turn out to be 'sells' after closer examination, and these are the stocks you'll find in this series. (I continue to add more each week as I find them.)

My current standard to declare a stock a 'sell' is that if the 10-year forward return expectations are lower than a 4% CAGR. All of the stocks in this article will have had a 10-year CAGR expectation of less than 4% at the time I wrote about them. You can find links to the original articles on my profile page here. (Type the ticker symbol into the "filter by ticker" box and it will pull up the articles I've written on that ticker. You will need a SA Premium subscription to read many of them, but you can check the publication dates if you'd like to double-check my work or see my sentiment rating at the time.)

The third step after one sells is to decide what to do with the proceeds of the sale. I call this one's 'default position', which is the place money sits while it is waiting to be invested in individual stocks. It is often assumed that a cash equivalent is the default position, but I actually prefer to stay invested unless it is clear we are headed into a recession very soon. I don't think a recession is imminent, but I do think we are late in the business cycle, so my suggestion is to use two 'defensive' positions as one's default positions right now.

From April, when I started covering the stocks in this series, until a few months ago, my suggestion for a defensive default position was a 50/50 mix of Invesco S&P 500 Low-Volatility ETF, and Invesco S&P MidCap Low Volatility ETF (XMLV), so that will be the assumed default position for most of the stocks I cover in this series. A few months ago I changed XMLV to an S&P 500 Equal Weight ETF (RSP). In an actual portfolio, one would have sold whatever XMLV position they had and bought RSP, but since that's too hard for me to track changes like that with standard SA articles, you will notice that a few of the more recent 'sell' articles I've written suggest RSP instead. So, for tracking purposes, I'll track whatever the original suggestion was without changes along the way.

Alright, now we identified a quality business, sold the stock because it was expensive, and put our money in either a 50/50 mix of SPLV/XMLV or SPLV/RSP. The next step is rotating back into the target stock and gaining free shares. For the simple mixed approach, there are two ways we go about buying back the original stock. The first way is to rotate back in whenever a 20-25% free share gain presents itself. And the second way is to rotate back in whenever the expected 10-year forward CAGR expectation reaches 8%, which I consider the long-term market average. I'm going to use a mix of both ways in this series as a way to demonstrate how they work.

Now let's examine how this strategy is working so far through the end of the month of November. I'll begin by examining the free-share-gain winners and any additions we had during the past month. Then I'll examine any of the CAGR-expectation winners. After that, I'll examine a variety of different free-share-gain categories before aggregating the data and returns at the end.

Free Share Gain Winners

As of the end of last month, we had five successful free-share-gain winners: Rollins (ROL), CSX (CSX), Fair Isaac Corporation (FICO) Union Pacific (UNP) and Norfolk Southern (NSC). This month we add one more: Expedia (EXPE).

I wrote a follow-up article on Expedia after the rotation opportunity came earlier this month if you would like to read the full article. Here is the performance from the time of my sell article until the free share gain goals were accomplished.

ChartData by YCharts

This was enough to produce about a +34% free share gain in Expedia using the free share gain approach. With that rotation, Expedia joins the other five successful free share gain rotational trades we've had so far:

Ticker Free Share Gains
ROL 41%
CSX 21%
FICO 23%
UNP 25%
NSC 21%
EXPE 34%

So, through the end of November, for the free-share-gain part of our simple mixed rotational strategy in which we aim to rotate back into the target stock after a 20-25% free-share-gain, we have had six successful rotations.

Now let's examine the other part of the rotational strategy of using a fair value 10-year CAGR expectation as a trigger to rotate back into the target stocks.

Full-Cycle CAGR Analysis

Through the end of October, of the 32 stocks we are currently tracking in the sentiment cycle series, only one, Union Pacific, had met the 10-year CAGR expectation of 8% and completed the second half of the rotation. Expedia joined Union Pacific this past month on November 13th, which I noted in the comment section of my follow-up Expedia article:

ChartData by YCharts

Rotating XMLV portion of the rotation back into Expedia when it hit an 8% expected 10-year CAGR on 11/13/19 would have produced about a +41% free share gain and fully completed the rotational strategy for Expedia.

Here are the two full-cycle CAGR winners we've had so far.

Ticker Free Share Gains from Full-Cycle Analysis
UNP 19%
EXPE 41%

So, overall, we've now had 8 successful, completed rotations out of 64 possible ones. Let's examine the 56 we are still tracking and see how they are performing. The following charts all run from the date of publication of the 'sell' articles through the end of November.

Stocks that are close to rotation (greater than +10% share gain)

Rollins 4/22/19

ChartData by YCharts

For the second half of the Rollins rotation, we are waiting for Rollins to offer an expected 8% 10-year CAGR. Since Rollins is currently trading at a PE just under 50, I think it's going to be a while before we are anywhere close to our goal. If one were to rotate back in today, though, they would have about a 30% share gain.

CSX 6/13/19

ChartData by YCharts

I examined CSX back at the end of September to see if it was near that 8% CAGR expectation and it wasn't quite there, yet. Since the price is higher today, I'm not going to run the numbers again this month. If was one to rotate back in today, they would gain about 12% more shares.

McDonald's (MCD) 7/16/19

ChartData by YCharts

For McDonald's, we are still waiting for a 20% free share gain opportunity. Right now we could gain about 12%, so I'll be keeping an eye on this one this month.

Intuit (INTU) 9/5/19

ChartData by YCharts

With Intuit, we are also waiting for a 20% free share gain opportunity. Right now we could get about 15%. I think there is a pretty good chance we could complete a rotation with this one at some point over the next couple of months.

Overall, we had 5 positions that fell into the +10% to +20% share gain category, and 1 that we are still tracking (Rollins) that is over +20%.

Stocks currently with a +5% to +10% share gain

Estee Lauder (EL) 8/19/19 (original 4/30/19)

ChartData by YCharts

Estee Lauder improved a category from the end of last month after the price rallied a bit.

Home Depot (HD) 11/18/19 original 6/28/19

ChartData by YCharts

Occasionally, I will perform an analysis on a stock and it won't quite be below my 'sell' threshold, but it will be close. Sometimes in those articles, I will note that if the price continues to rise a certain amount, then it will become a 'sell' in the future. That is what happened with Estee Lauder last month, and this month, it happened with Home Depot. Here is what I had to say in my original Home Depot article:

For those who own the stock, whether to sell or not is a tougher call. Technically, Home Depot is just in the "underperform" category and not a sell yet. There isn't a ton of easy value to be found in the market right now, especially among large-cap stocks, and yields on treasuries are low. The main considerations I would point to is that Home Depot's earnings might be more cyclical than some investors realized. There are lots of investors who might be surprised to see Home Depot's earnings fall -30% or -40% during the next downturn. If you are a long-term holder, you should be aware that this sort of earnings decline is possible and even likely during the next recession. This isn't a consumer staple stock that only suffers single digit earnings declines during downturns. Additionally, if the stock price were to rise another 15% from here rather quickly, the higher valuation would certainly move it into the 'sell' category. So, if I was on the fence, and got a big pop in the price over the next few months, that should be enough to move one off of the fence into the 'sell' camp.

Home Depot hit that 15% threshold right before its last earnings report, which disappointed the market, and so it joined the list of stocks we are tracking.

Norfolk Southern 6/12/19

ChartData by YCharts

You will recall that we already rotated half of the Norfolk Southern idea back in for a good gain. Since that time, the stock has rallied a lot. We are still waiting for the price to fall far enough to achieve and 8% expected 10-year CAGR before rotating the other half back in. Right now, I'd estimate the price needs to drop about -20% before that happens.

Starbucks (SBUX) 9/13/19

ChartData by YCharts

Starbucks has been relatively weak since I wrote about it in September. I expect it's only a matter of time before we get some solid free share gains with this one.

Stryker (SYK) 10/17/19

ChartData by YCharts

Soon after I covered Stryker in October the company announced an acquisition that sent the stock lower. It still has quite a bit farther to fall before we would rotate back in, though.

Overall, we had 9 positions that fell into the +5% to +10% share gain category.

Stocks that aren't doing much (-5% to +5% share gain)

About half of the overall positions, 29, fell in this middle category. We would expect this to be a large category, but this month is especially big because these ideas tend to take several months to play out, and I've been adding lots of stocks to the list the past couple of months. This past month, I think I added 8 stocks, and since each sell idea is split in two, that represents 16 new positions added in just the past month or so. As time goes on, I expect many of these to move more than this 10% range relative to the default alternatives I suggested, but for now, there are a lot that aren't doing much yet. I'll give limited commentary on these and just post the charts with the dates listed for most of them.

Sherwin-Williams (SHW) 11/27/19

ChartData by YCharts

Walmart (WMT) 11/19/19

ChartData by YCharts

Fair Isaac Corporation 8/7/19

ChartData by YCharts

Fair Isaac is pretty interesting because we rotated half of our rotation back in already, and the stock has rallied over 20% since then and is almost back to even with the default position. This serves as a good example of the differences between the two rotational approaches we are mixing together. Often, simply aiming for a 20-25% free share gain is easier to do than waiting for an overvalued stock to fall all the way to fair value. The trade-off is that if a stock is really overvalued, often it can keep falling after that 20-25% threshold has been met, and one's free share gains are limited to that 20-25%. It's not always the case that stock immediately bounces back as FICO has.

Accenture (ACN) 9/12/19

ChartData by YCharts

Automatic Data Processing (ADP) 4/25/19

ChartData by YCharts

Brown-Forman (BF.B) 11/8/19

ChartData by YCharts

CGI (GIB) 8/28/19

ChartData by YCharts

Hershey (HSY) 9/11/19

ChartData by YCharts

Illinois Tool Works (ITW) 10/24/19

ChartData by YCharts

Mastercard (MA) 10/9/19

ChartData by YCharts

Medtronic (MDT) 9/26/19

ChartData by YCharts

McCormick (MKC) 8/26/19

ChartData by YCharts

Nike (NKE) 9/18/19

ChartData by YCharts

Northrop Grumman (NOC) 10/23/19

ChartData by YCharts

Paychex (PAYX) 6/22/19

ChartData by YCharts

Teleflex (TFX) 10/16/19

ChartData by YCharts

Texas Instruments (TXN) 10/31/19

ChartData by YCharts

Stocks currently with a -5% to -10% share gain

You'll notice that a lot of the stocks in this category have been split between this category and some other category. That mostly has to do with the relatively poor performance of SPLV, one of the default positions. This is because SPLV itself had become overvalued over the past year or so. SPLV rebalances and reconstitutes itself before December, though, so I'm hoping that we get a little bit better mix in SPLV for the next quarter. This past quarter, it was heavily weighted toward utilities, which are quite overvalued right now.

Nike 9/18/19

ChartData by YCharts

Illinois Tool Works 10/24/19

ChartData by YCharts

Teleflex 10/16/19

ChartData by YCharts

ResMed (RMD) (9/23/19)

ChartData by YCharts

Mastercard 10/9/19

ChartData by YCharts

Lowe's (LOW) 7/15/19

ChartData by YCharts

Cintas (CTAS) 5/16/19

ChartData by YCharts

Cintas's price started to pull back after a short report last month. Personally, I tend to focus more on the basic valuation, and I do still think it's too expensive.

Stocks that might have been misjudged: (currently -10% share gain or more)

ResMed (9/23/19)

ChartData by YCharts

Target (TGT) 11/18/19

ChartData by YCharts

The day after my Target 'sell' article came out they reported very good earnings and the stock had a big pop. This happens sometimes. I'm taking a medium-term view of 2-3 years with all the stocks in this series. Sometimes I get lucky with my estimates, like with Home Depot, and the price drops right after it crosses my 'sell' threshold. And other times earnings come out and the stock has a big pop like Target. I think Target had a little more potential for it since its P/E wasn't as high as many peers. All they have to do get the stock price to rise is find a way to get some earnings growth. This has a higher probability of happening than with a stock that already has a very high P/E. In those cases, they can still have excellent earnings growth sometimes and not have it enough to push the stock price higher.

This month there ended up being 3 positions that fell in the -10% to -20% category. There were no stocks in the beyond -20% category.

Compiling ongoing results

There are several things I am attempting to test in this series. The first is whether one can successfully determine a good time to sell a high-quality stock. The second is whether one can successfully find a suitable alternative for that money. And the third is whether one can successfully determine a good time to purchase the stock back. My hypothesis is that it is possible to consistently do this with certain high-quality stocks and gain 20-25% worth of free shares in the process.

Right now we have 32 stocks we are tracking and each one of those was divided into two, so we have 64 total positions. I'm going to keep adding to this list as long as I can find high-quality, yet expensive, large-cap stocks to write 'sell' articles on. I want as many data points as I can find. Usually, there are a few outliers that have something good or bad happen to them and I want to capture those in the series if I can, so that we can get a full look at the range of possibilities.

Source: Author

The chart above shows the current distribution of 64 positions relative to the amount of free shares gains one would have had at the end of November. If the results were random one would expect a normal bell curve. For the bulk of the positions, that is in fact what we see this month. The one part of the distribution where that is not true is the tail ends. On the far positive +20% free share gain end, we have 8 winners, while on the far negative -20% free share gain end, we 0 losers. It's already getting difficult to make the case that the distribution could be explained by randomness, and I expect that over time it will get more difficult.

As I noted earlier, several of the underperforming positions can actually be explained by the underperformance of SPLV. You probably noticed that I included SPY as a reference point in the charts as well so that we could know how one would have done simply using SPY as the default position and we could measure my choice of default positions separately from the performance of the target stocks. Below is the same chart distribution of the sell ideas only this time versus SPY instead of my default choices.

Source: Author

You can see the clear skew to the left in this case. This allows us to measure both the 'sell' portion of my idea, and also measure the usefulness of my default position suggestions. What we have seen so far is that my 'sell' suggestions have been much better than average when we use the S&P 500 as the default position. And while my default suggestions have been doing well relative to the 'sell' suggestions, they haven't done as well as the S&P 500, yet. This could change if there is a market correction, though, so I wouldn't draw any firm conclusions just yet.

Conclusion

So far, this series is off to a great start. We've had 8 successful rotations and no total disasters. I will add more stocks to the series this month in an effort to warn as many owners of overvalued stocks as possible about the dangers of low future returns, and to get as large of a sample size as possible for this real-time experiment. If you would like to follow along, make sure to click the orange 'Follow' button at the top of the page to be notified when I publish new articles.

Disclosure: I am/we are long RSP, SPLV. I wrote this article myself, and it expresses my own opinions. I am not receiving compensation for it (other than from Seeking Alpha). I have no business relationship with any company whose stock is mentioned in this article.