Last week's article, High Yield Or High Growth: Which Belongs In Your Portfolio?, presented a handful of spreadsheets that calculated the future results for varying initial dividend yields and growth rates. However, while those tables are helpful in showing potential results that differing growth rates can provide, they fall short of providing actual real-world analysis. It's one thing to have a mathematical formula produce results, but quite another for a stock to do the same.
In that spirit, I thought it would be beneficial to find some companies that have produced long track records of earnings and dividend growth to see if their returns came close to matching the tables' forecasts.
To do so, I will be using F.A.S.T. Graphs, which does an excellent job of showing the historical valuation for stocks, as well as the all-important earnings growth, dividend growth, beginning dividend yield, cumulative dividends, and total returns over varying time periods. Essentially, all of the information that was presented in my theoretical return matrix.
For this exercise, I will identify companies from the varying growth and yield areas: high yield/low growth, mid-yield/mid-growth, and low yield/high growth.
F.A.S.T. Graphs has data going back to 1998, and is able to provide a snapshot of varying time frames over the spanning years. For my comparison, I will look for a rolling 10-year period from which the company began and ended trading at a similar valuation level. I will then compare how the stock performed over that decade with what the tables project for a similar yield and growth rate.
The first example comes from a company that calls itself "The Monthly Dividend Company", Realty Income Corporation (NYSE:O). Realty Income is known as one of the most consistent and reliable REITs in the market, and is widely held by dividend growth investors.
For this stock, I was able to find a period from 12/31/2003 to 12/31/2013 where it began trading at a forward yield of 6.3%, grew dividends at a 6.3% annual rate, and grew FFO at a 5.1% rate.
From the graph, you can see that the company traded below the fair value trend line for roughly the first half of the decade and then above the line after the recession.
First, here are the total returns with dividends being collected.
For comparison's sake, here are the 10YR projection tables:
Looking at the % YOC comparison, you can see that the table projects a 10.7% YOC for a 6% yield with 6% growth rate (highlighted in green on left table), which compares very closely with the 10.9% that was provided by Realty Income. Similarly, the total return of 169% [($26,871-$10,000)/$10,000)] comes close to the 158% projected, and the cumulative dividends of $8,206.20 comes close to the projected $7,900. In each case, Realty Income's returns slightly beat the 6%/6% projection, which makes sense considering its actual 6.3%/6.3% rates.
Now, here are the results with dividends reinvested:
Again, you can see that the tables and actual results correlate quite nicely. Realty Income produced a YOC of 18.9% compared with the projected 18.6% and provided a total return of 231% compared with the projected 221%. Also, note that the total return rate of 12.7% nearly matches the initial yield plus growth rate of 12.6%
The next higher yield example is HCP, Inc. (NYSE:HCP). HCP is a healthcare REIT with a 31-year streak of dividend increases, although I will point out that the stock is under some pressure right now due to concerns with its HCR ManorCare portfolio. That said, it still remains a good example to use against the returns tables.
For HCP, we will look at the 12/31/2004 to 12/31/2014 time frame. Here, you can see the company grew FFO at a 6.2% annual rate, began the period trading at a P/FFO of 16.7 and ended at a P/FFO of 14.5.
Now looking at dividend growth, you can see the company grew dividends at a 2.7% annual rate during the period. This resulted in a total return of 122.7%, a YOC of 7.9%, and a cumulative income return of 68.2%.
Now look at the projection matrices to see how the real world compares:
Using a 6% initial yield and 3% dividend growth without dividend reinvestment, you get a projected YOC of 8.1%, total returns of 103%, and cumulative income return of 69%. These numbers compare favorably to the actual results of 122.7%, 7.9%, and 68.2%, respectively. The difference in total returns is understandable as HCP grew FFO at a 6.2% rate, twice the 3% rate from the table, which led to higher capital gains returns, despite the change in P/FFO during the period.
Now looking at the results with dividend reinvestment, you can see that HCP provided a YOC of 13.5% and total returns of 178.2% compared with the matrix prediction of a 14.2% YOC and 141% total returns.
The results are reasonably close, and it's easy to see where the difference lies. If you look at the payout ratio column, the dividend payout dropped from 89% of FFO at the beginning to 72% at the end, which resulted in a declining yield as time progressed. Since dividends were being reinvested at lower and lower yields, the actual YOC ended up lower than the table's projection, which assumes a constant yield during the period. Total returns are higher than the projection for the same reason they were higher in the first comparison; FFO grew at twice the rate of dividends.
Our next example is Reynolds American, Inc. (NYSE:RAI), a tobacco company that has been one of the top dividend growth stocks in the market. It seems hard to believe when looking at today's prices, but RAI actually used to be a high-yield stock. Back in the early 2000s, the stock routinely traded with yields in the high-single digits, and for this example, it started at an initial yield of 6.6%.
The company grew EPS at a 9.7% rate and dividends at 10.1% while also experiencing an expansion of its P/E multiple from 12.3 to 14.7. This resulted in total returns of 295.7%, a YOC of 15.3%, and total income return of 101.4%.
This compares with projected returns (6.5% yield @ 10% growth) of 263%, YOC of 16.9% and income return of 104%. The higher-than-expected total returns would be expected, as Reynolds' P/E multiple expanded from 12.3 to 14.7, resulting in higher capital gains. The lower YOC is a bit more puzzling, however, I expect that is a result of the bumpy dividend increases rather than a linear growth rate, as real-world increases ranged from 0% to 31% compared with the 10% annual increases used for the tables.
The reinvestment case also matches up fairly well, with a YOC of 28.5% compared to 29.9% and total returns of 460% compared to the 387% projected. Also, note that the annualized total returns of 18.8% exceed the growth plus initial yield of 16.5%. These differences can again be explained by multiple expansion and a bumpy dividend growth rate.
The final higher-yield example comes from AT&T, Inc. (NYSE:T), which is another stock widely held by dividend growth investors. AT&T ended 2005 at a 5.5% yield, and over the course of the next decade, grew EPS at a 4.7% rate while increasing the dividend payout at a 3.8% rate. It also saw its P/E multiple contract from 13.8 to 12.7.
During the period, AT&T produced a total return of 109%, a YOC of 7.7% and an income return of 68.7%.
This pretty closely matches the projected numbers of 114%, 8.1%, and 66% for dividends not reinvested.
The total returns with reinvestment of 137% lagged the projected 153% return and the YOC of 12.8% lagged the 13.6% as well. However, this should be expected, as the actual growth rate of 3.7% was lower than the 4% in the table, and there was also a large spike in share price in 2006 and 2007 that lowered the yield at which reinvestment occurred. Additionally, the stock traded at a lower P/E multiple at the end, which lowered capital appreciation.
Enbridge had a great run from 2003 to 2013, during which it grew its payout at a 14% annual rate and its cash flow at 13.1%. Its P/CFL multiple also expanded from 9.5 to 11.7, which will come into play below.
Combined with an initial yield of 3.5%, this led an impressive total return of 393%, YOC of 11.5% and income return of 70%.
This compares favorably to the matrix returns of 338%, 13.0% and 68%. Again, the outperformance in total returns is the result of multiple expansion over the period while the YOC lagged because of the dividend cut in 2008.
The actual reinvested total returns of 478% also outpaced the projected 423% while the 15.5% YOC lagged the 17.6% that was projected. Again, those are both explained by multiple expansion and the bumpy dividend growth rate.
Next up is Avista Corp. (NYSE:AVA), which is an electric and natural gas company located in Spokane, WA. Avista is a bit different case than the previous ones as it saw an expanding payout ratio during the period while also seeing a fair amount of volatility in earnings and share price. It also saw a decline in its P/E multiple from 19.7 to 15.2 during the period. Let's see how much of an impact those variances had on returns.
Here you can see that dividend growth varied widely from 4% to 23% and also how the payout ratio varied from 39% to 88% due to the volatile earnings. Overall, Avista provided a total return of 101%, a YOC of 6.7% and a dividend return of 45% during the period.
This compares with the 182% total returns, 7.1% YOC and 46% dividend income from the tables. The lagging YOC is understandable considering the initial yield of 2.8% was below the 3% from the table, and the total returns being lower were caused by the previously mentioned P/E contraction and the lower EPS growth rate.
Similarly, the total returns with reinvested dividends of 125% significantly lagged the projected 218% returns. However, the declining multiple and corresponding higher yields that shares were reinvested at during the decade actually led to an outperformance in the YOC at 9.6% compared with the 9.3% projection.
Avista is an interesting comparison against the matrix. While investors gave up some capital gains due to a compressing P/E ratio, they were able to gain an income advantage by picking up more shares through reinvestment.
The next company, Hormel Foods Corp. (NYSE:HRL), is best known for its signature meat product, SPAM. However, it is also well known as a top dividend growth stock, as it owns a 50-year streak of dividend increases.
Hormel has been a consistent performer, and had a nice stretch from 2000 to 2010 where it grew EPS and the dividend at right around 9% annually. It also saw its P/E multiple expand by about 12% during the period, from 13.5 to 15.1.
This resulted in some excellent returns, especially when compared to the negative returns from the S&P during the decade. Hormel produced total returns of 206%, a YOC of 4.5% and a dividend return of 32%.
This compares with the projected returns of 167%, 4.7% and 30%. Again, the income components were close, yet the total return outperformed due to multiple expansion and a slightly higher than 9% EPS growth rate.
The comparison for reinvestment was similar as the actual total returns of 225% exceeded the 189% from the tables while the YOC of 5.3% lagged the 5.7% that was projected.
The final set of comparisons will look at two high-growth companies that started with 2% or lower initial yields.
First is Lockheed Martin (NYSE:LMT), an aerospace and defense company headquartered in Bethesda, MD. LMT is an interesting case for this article, as it had a wide swing in valuations as it traded at a P/E range from 8 to 17, and also dividend yields from roughly 1.5% to 4.5%. It also produced a 14.9% EPS growth rate for the decade despite seeing earnings stagnate between 2008 and 2012 and saw its payout ratio double from the mid-20% to high-40% range.
Despite all of the ups and downs, those who held through it all did quite well. Those who bought at the end of 2004 and held through 2014 saw total returns of 298%, a YOC of 9.9% and a cumulative dividend return of 51%.
For the matrix, I have highlighted the 2% yield and 19% growth rate for the dividend portions and the 2% yield with 15% growth for the total return portions. With a lower-yielding stock, capital gains will outweigh dividend returns, and considering the EPS growth lagged dividend growth, I felt this was the better comparison.
Based on those numbers, the actual YOC of 9.9% lagged the projected 11.4% while the cumulative income slightly outpaced the 49% projection. Total returns were also lower than the 345% projection, due to P/E multiple compression.
The reinvested results were a bit closer however, as total returns were 363% compared with the 393% projected while the YOC was 13.1% compared with the 13.5% projection.
When it comes to a company with as many variable as Lockheed Martin, the tables weren't quite as effective in their projections. While the dividend comparisons were fairly close, the total returns lagged, even when using the 15% growth rather than 19%.
The final example to look at is Casey's General Stores, Inc. (NASDAQ:CASY), a convenience store chain that is a lesser-known dividend-growth company, albeit one that has been a terrific performer over the years. It produced both EPS and dividend growth of over 18% during the 2005-2015 time frame, and its consistent growth should make it a good comparison test for the matrix.
As one would expect, with an 18% growth rate, Casey's rewarded investors some tremendous returns during the period. It produced a total return of 413%, a YOC of 4.3% and a cumulative income return of 26%. Its 17.7% annualized return beat the S&P by more than 10%!
The total returns lagged the projected returns of 447% due to a PE multiple that fell from 20.5 to 17.8. The YOC also fell short of the projected 5.1% number. I suspect this is due to a difference in timing between the chart and the table, as my projection table matrix has the growth rate compounded to the 10th power while the F.A.S.T. Graph goes to the 9th. This is likely the cause for the YOC difference with Hormel as well. However, the cumulative income was slightly higher than projected, likely due to the initially higher yield of 1.1% compared with the 1% in the table.
The reinvested returns of 440% and YOC of 4.7% also lagged the projected 478% and 5.7% numbers, for the same reasons noted in the base case.
Thoughts & Observations
Overall, I think the projections and real-world results compared quite well. Where there were differences between the two, it was fairly easy to point out what caused the differences.
Here are a few observations I've made while putting this all together:
- Buying at a proper valuation is incredibly important in maximizing your long-term results. Not only do you lock in the initial yield forever, but also the compounding effect on that yield magnifies that difference over time. Additionally, if you overpay, you lose out on potential capital gains with P/E normalization as the stock returns to fair value in the future.
- The projection matrices work best for stocks that are fairly consistent in their earnings and dividend growth. Volatile swings in share price, dividend growth, and earnings growth will cause a divergence from the straight-line growth projections.
- These projections assume a constant payout ratio and will not correlate as well to a company that is rapidly increasing the dividend while growing EPS at a lower rate.
- Yield and growth are not necessarily mutually exclusive, and total returns can come from either. Reynolds American was able to grow an initial yield of over 6% at a double-digit rate that produced annual returns of around 18%. Meanwhile, Casey's had an initial yield of just 1% yet matched the returns, albeit with much lower income.
I hope this was helpful in showing how the projection tables work and how they should be looked at in comparison to real-world investments. In the third part of this series, I will identify a list of companies that fit into the different growth/yield categories and attempt to project future returns for each of them.
Disclosure: I am/we are long O, LMT, RAI, T.
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.
Additional disclosure: I am a Civil Engineer by trade and am not a professional investment adviser or financial analyst. This article is not an endorsement for the stocks mentioned. Please perform your own due diligence before you decide to trade any securities or other products.