Q1 2018 Quarterly Letter

Download the Quarterly Letter

Download

2018 and Beyond

The next 10 years will be about getting the return portion of risk allocation right.  

• The largest contributor to total return will be dividends due to low/modest GDP growth.
• Traditional fixed income will only reduce volatility, but will not meet investors income or return needs.
• Combining stocks and bonds with liquid alternatives will deliver returns to the client that will meet their long‐term investment needs at a level of volatility they can withstand.   

Historically, we have devised investment strategies that we believed would be appropriate for three years into the future. We have told our RIA clients that seeing beyond three years is very cloudy. Today, however, we find ourselves looking into a future that appears imminently cloudy. The next three or five years look very uncertain with variables that will make investment outcomes widely variant depending on how economic issues and market corrections develop. So, let’s explore what we mean.

It is difficult to read a discussion on the future returns of the stock market without reading about the coming correction. Stock prices have risen since the market hit bottom in March of 2009, and have been virtually uninterrupted since then. The theory of the correction is based on history; markets do not go straight up and the longer we go without a correction, the more likely there will be one. Well, that is a reasonable theory and hard to suggest the logic is impaired. The problem is determining when it will happen and how it will happen. Many market prognosticators have been talking about the 2018 correction, and we may have that correction in 2018. But, we ask, what if we have the correction in 2019 or 2020 or even 2023? The reality is corrections have been impossible to predict and come in so many forms that they actually seem like they are not related. So, let’s look at three corrections: 1987, 2000‐2002 and 2008.

1987

October 19, 1987 is referred to as Black Monday when the Dow Jones Industrial Average fell 22.6%. So, does that mean that between the market opening and closing, investors determined rationally that stock valuations as of the close of Friday, October 16, 1987 were overvalued by 22.6%? Technically, the answer is yes, but there were factors that exaggerated the trading and sell orders were submitted without regard for the price of securities.  Overvaluation and program trading schemes were the reasons most commonly used to explain this decline. This day followed declines of 3.8%, 2.4% and 4.6% on the preceding trading days, Wednesday through Friday, October 14 – 16. What this meant for investors was that $1.00 invested in the Dow Jones Industrial Average before the opening on Wednesday, October 14, 1987 was worth 69 cents after the close on Monday, October 19, 1987. So, investors lost 31% in 4 trading days. We had never seen such a short‐term decline, and we have not since. 

2000-2002

The tech bubble of the late 1990’s peaked in March of 2000 and stock prices fell for almost three years. The S&P 500 returns for those years were as follows: 

2000: ‐9.1%

2001: ‐11.9%

2002: ‐22.1% 

The total decline was 37.6%. $1.00 invested on the first day of 2000 was worth $0.63 at the end of 2002. 

So, the first observation one would make when comparing these two declines is that they were similar in order of magnitude: 31% in 1987 and 37% in 2000‐2002. However, they were dramatically different in terms of length, four days in 1987 and almost three years in 2000‐2002. 

There is one more observation we would make on the 2000‐2002 correction: it was not universal. While the S&P 500 fell by 37%, our portfolios consisting of high quality, high cash flow, growing dividend stocks were up over 15% during this period. There were other thoughtful investment portfolios that also protected capital during this three‐year decline. 

2008

The stock market crash of 2008 was quite different from the 1987 experience or the 2000‐2002 experience. Largely this was because 2008 started the largest recession we have seen in this country. Stock prices started to decline in late 2007 and bottomed in March 2009 and the total loss was just over 50%.  

There was no place to hide in 2008. Everything declined: stocks, high yield bonds, preferred stocks and commodities. The S&P 500 was down over 37% in 2008, but dividend stocks were down only 25%, so there was an order of magnitude difference. But a decline of 25% is enormous by anyone’s measure. 

The Three Corrections 

What can we learn from these corrections? 1987 was very short, less than one week, while 2000‐2002 lasted almost 3 years. High quality stocks protected investors and actually provided positive returns in 2000‐2002, while they gave limited protection in 2008. The following three generalities can be said about corrections. If you had studied stock market corrections as we have, you would find these three guidelines are applicable for every correction.

1. They can be long or short.
2. Sometimes there will be a place to get positive returns but sometimes not.
3. Any correction connected to a recession will be severe.

There is one more observation we would mention on corrections. They cannot be predicted! If you read  the financial press from early October of 1987, late 1999 or the fall of 2007, you will find very little, if any commentary, on the coming stock market correction. What is true is that television commentators will find some pundits who predicted these corrections, and this will make a great story, but what they will not tell you is the lack of reliability of their so‐called experts.  

What all of this says to us is that we should have an equity investment thesis that will survive the next correction and a rationale for not only living through it but also providing confidence that your portfolio will succeed when that happens. That equity investment thesis is a growing dividend model that allows us to see growth in dividends when living through even the worst declines.

So, the inevitable and unpredictable nature of corrections is what clouds our view of the next three to five years. We think most likely there will be a correction, but whether that is the 2018 or 2023 correction, we have no idea. 

10 Years

10‐year stock market predictions have two historical characteristics. They will have more guesswork than we would like, but they are statistically more accurate than three‐year predictions. The reason for the accuracy of 10 years being greater than three years is easy to understand; the impact of a correction on three years will always be greater than that same impact on 10 years. Any three‐year period that includes 2008 will have abysmal returns, and any three‐year period that starts after 2008 will look great. So, this is why we find ourselves looking out farther. 

What do we use as we look out 10 years? GDP growth has been the most reliable predictor of long‐term stock returns. Like all countries, we go through periods of high and low GDP growth periods. 

GDP Growth

Chart 1 shows the GDP growth in this country from the end of World War II to the present. As you will see, we have shown two distinct periods, 1947‐1999 and 2000‐2017. So why did we have high growth from 1947‐1999 and low growth over the last 18 years? Let’s examine the two periods. 

Chart 1Source: Bureau of Economic Analysis 

1946‐1964: The Baby Boom

World War II ended in 1945 and our country developed many projects to push for faster economic growth. Women left the factories and men took their place. We went from making bombs and guns to toasters and appliances. We went from a war economy to a consuming economy as the baby boomers were born. Government spending led the early years, but private spending replaced government spending in the later years, and our economy boomed. We built new highways, and suburbs were created with new houses, wider lawns and a sense of freedom from crowded cities. We had the GI Bill which sent unprecedented numbers of people to college. 

Baby boomers are those born between 1946 and 1964, so the need for more spending persisted.  The economic needs of expanding families contributed for over 20 years after the end of World War II. 

But like all periods of economic expansion, inflation caught up with us as we came to the end of the 1960’s. While GDP growth continued to be very positive, inflation contributed mightily to those numbers. We saw a stock market that stagnated from 1969‐1981; the S&P 500 compounded at 5%, but 4% was from dividends and only 1% was stock price appreciation. 

1982‐1999: Baby Boomers Come of Age, The Tech Boom and The Rise of China  

1982 marked the beginning of the largest economic expansion in our country’s history. The baby boomers were having babies and while family size was declining, the base of baby boomers was enormous. This led to the need for more toasters, ovens and dishwashers. It also meant more cars as we became a nation of 2‐car families. So, the economic push of the families of baby boomers mirrored the impact they had at birth. Companies expanded as the demand for their products expanded, and life was good in the American economy. 

The factor cited by economists that contributes the most to economic expansion once we have reached full employment is productivity gains. If businesses can produce more widgets with the same employees in the same time frame, they can have larger profits. The 1980’s and the 1990’s saw technological advances at a pace never seen before, so productivity gains were the second factor that our economy witnessed in this period. 

China was the third factor. China is a country with approximately four times the population we have in the U.S. So, in the 1980’s and 1990’s when the population went from the farm to the factory, China needed more of our goods and services, and our economy enjoyed huge numbers of new customers.  

2000-2017

The three factors that gave us the greatest bull market in our country’s history saw indications of maturing over the last 18 years. 

The first baby boomers reached 65 in 2011. 2017 saw the 7th group reach that age, and we have 12 more groups to go. When this group were in their peak spending years, 30s and 40s, their spending patterns boosted our economy. Today they are in their 50s, 60s and 70, the saving years, and saving does not boost an economy. 

The productivity changes from technology continue, and this will add to our growth. We read today about AI, autonomous cars, speed and storage computing capacities, and all of this will help us. But it is unlikely to add to productivity in the way their predecessors did in the 1980s and 1990s. 

China is huge, but it is slowing. They have indicated they want to go from a producing to a consuming economy, so they will continue to be a great partner, just not at the same growth rate.  

Our view on the slowing of GDP growth from 3.5% to 2.0% is that it will continue. We would be very happy if we saw higher GDP growth, but higher GDP growth requires catalysts just like the ones we had after World War II and from 1982–1999. Catalysts cannot always be seen in advance, so we can react when we see them. But planning on them and then being disappointed is not wise. Prudence demands we plan for all outcomes, even the ones that may not be the most desired. 

So What Works in Low GDP Periods? 

This begs the question of what works in low GDP growth periods. We have two periods that will give us some guidance, 1969‐1981 and 2000‐2017. 

The first period was a time of high inflation, but stock returns were muted. GDP was higher than the last 18 years, but inflation adjusted GDP was not. The S&P 500 compounded at 5.6%. The last 18 years saw the S&P 500 compound at 5.1%. 

There are two factors that make up these returns, stock price appreciation and dividends. For the first period of 1969‐1981, appreciation was 1.3% and dividends 4.3%, while the last 18 years the break down was 3.4% appreciation and 2.0% dividends. 

There are likely any number of ways to improve on these 5% return periods. You will read about better stock selection schemes and some market timing ideas. We think focusing on greater appreciation comes with greater risk and less predictability. We advise focusing on dividend level and dividend growth. Jeremy Seigel reconfigured the S&P 500 based on ranking stocks by their contribution to the dividends of the index. His work showed that during the 1969‐1981, dividend strategies produced 9.2% returns as compared with the index return of 5.6%. During the last 18 years, dividends produced 8.6% against the index of 5.4%. So, our rhetorical question is if we can improve returns by 60% by focusing on dividends, why risk another strategy that may be volatile? To possibly get higher returns? We like the safety of our idea. 

You may be aware that we have built growing dividend strategies with world class managers and have written a white paper titled “The Case for High & Growing Dividends”, which we would be happy to share with you. The current economic environment confirms our decision to continue to follow that advise. 

What if we are wrong? What if GDP growth is high? 

While we think the catalysts do not exist for high GDP growth, there is always a possibility that factors we cannot see now play out in a way where higher GDP growth occurs. We have three thoughts on this topic.  

Point number one relates to Jeremy Siegel’s study of the reconfiguration of the S&P 500 based on an emphasis of dividends.  His study shows that from 1982‐1999 the index outperformed dividends, 18.5% to 17.6%, but for the first 16 years of this 18‐year period, dividends actually outperformed the index, 17.8% to 16.6%.  

Point number two relates to the current investment climate. Stocks prices are at least fairly priced, some think overpriced. Whichever camp you agree with, it is impossible to believe stock prices are a bargain. The dividend yield of the S&P 500 is just 2%, so in order to get a reasonable return from here by investing in the index one would need high levels of price appreciation. We feel much more comfortable with a conservative strategy and a current dividend yield of 3.8%. 

Point number three relates to the bond market. The bond market’s very low yields may be telling us something about the stock market. There is no statistical data to support a thesis that low bond yields mean prospective low GDP growth, but we should not ignore it. When the Fed lowered rates after the depression, it took a war, 25 years and a fair amount of stimulation to get the economy on track.  

Today, many pundits are suggesting that the new lower tax rates corporations will be paying will be a stimulus, and we agree it will be. The question is how much of a stimulus. We think not enough to push us in a sustaining way over 3%. We may have a quarter or two of 3% GDP growth, we may even have a year, but there will need to be other catalysts to get us over 3% in a sustainable way. So, we think we have history on our side, and we can be flexible. We think our dividend strategies will give us returns that will fulfill the investment needs of your clients.  

10 Years 

So, what do we see 10 years from now with equity predictions? We think we will have lived through the 2018 or 2023 correction. We think we will continue to have 2% or 2.5% GDP growth as our population and Europe’s population age. We think the largest contributor to total return will be dividends. Today our portfolio using domestic large and small cap manager and an international manager with 15% in small cap and equally divided between domestic and international yields 3.8%. Historical dividend growth for the S&P 500 has been 7%; our portfolio has been higher. Now it is possible we will have lower or higher dividend growth. It makes sense that if we have 2% GDP growth, it is possible there will be pressure on dividend growth; yet todays companies’ payout a lower percentage of earnings or cash flow in dividends then they have historically. So, we will use 7% in our example. If we start with our portfolio yielding 3.8% and increase dividends by 7% annually, the compounded cash flow from those dividends would be 5.61% over the next 10 years. The last 18 years our equity portfolio annualized at 8.5%. Will it achieve that over the next 10 years? We think it is possible to assume that; we would need to get 2.9% from appreciation, well below the 7% long‐term appreciation of the S&P 500 and slightly below the 3.4% of the last 18 years. We should point out, however, that this 2.9% appreciation is well above the 1.4% from 1969‐1981.  

Now it is obvious to everyone that any prediction has a margin of error. We could be high on our dividend growth assumption and over on our appreciation return assumption. The 2018 correction could take place in 2027 wiping out the gains of the next 9 years. So, we express these opinions with these caveats in mind. 

The Silver Lining  

This still leaves the question of how RIAs should discuss all of this with their clients, and we think this change to a 10‐year view presents an interesting opportunity that will strengthen commitment and trust between you and your clients.

Let’s start with the client. They come to you because they want and need guidance. All of the research shows clients come to RIAs for two main reasons and the basic premise of these is they do not want to do their own investment research and planning. They come to you as their RIA because they believe you will know everything about their goals, fears, and desires, as well as their trust and tax issues. They also believe you would never recommend anything to them that is not in their best interest, you have their back.    

One of the common themes we hear from our RIA clients is that most of their clients fit this description and will follow their guidance. Others of their clients, however, watch too much television or read too much financial press. These clients want to know your opinion on the latest version of the greatest investment opportunity they heard from Cramer or some other impressionable television personality. These clients likely still follow your advice, but perhaps question it a bit too often. We think our analysis of the shift toward the long‐term may help you with these clients, and we have an example of one of the stocks in our portfolio that makes the case.  

Electric cars appear to be a big part of our future, and this is very encouraging on a number of fronts. They have fewer parts, require less maintenance, significantly reduce pollution. Tesla is the car company based on this idea, but other companies have a plan to produce large numbers of electric cars. Last year, BMW sold 100,000 electric cars. So, if our goal was to invest in this idea, which company would you buy? You could buy Tesla and it might be the leader 10 years from now. The price of the stock is up tenfold over the last five years. In addition, if you look at their financials, they do not have a P/E ratio because they have no earnings. It may also be possible that Tesla does not succeed, and the price instead of rising, falls. The history of buying stocks after being up ten times in five years is not good, actually that is an understatement. Tesla is priced for perfection, and it may succeed. The question investors should ask is not who will succeed but how do you find a prudent way to invest in this idea.  

It might be that Ford, General Motors, Volvo and BMW, just to name a few potential competitors of Tesla, will successfully compete in the electric car business and one of them or all of them slow or even wipe out Tesla. This is a “make it or break it” investment thesis.  

Our portfolio has a stock that will succeed whether Tesla wins or losses. Electric engines require semiconductors, lots of them. We own Taiwan Semi‐Conductor, and they will be happy to sell their semiconductors to Tesla, Ford, GM, Volvo, BMW or any company making battery operated cars. Today Taiwan Semi‐Conductor has a 2.9% yield and their dividends have grown from $0.39/share ten years ago to $1.16 today, annualized growth rate of 11.5%. We think this example of not looking for the winner, but looking for stable companies helping the latest technology innovate may help your small group of clients who get too many sound bites. The biggest winner in battery operated cars will not be Taiwan Semi‐Conductor. It will be Tesla or Ford or some other company we do not know just yet. But we do not think a prudent investment thesis should look for the “winner”. We think a prudent investment thesis should look for high quality, high cash flow growing businesses that will benefit from the technological advances we see. If Taiwan SemiConductor raises its dividend even half of what it has over the last 10 years, our clients will be very happy they have this company in their portfolio. In addition, the additional income will cushion their portfolio.  

Bonds

Interest rates are as low as they have even been in the history of our country. We were here once before. Just as the Federal Reserve Bank lowered Fed Funds rates to zero in 2008, they did the same after the 1929 depression. Rates stayed low through the 1940s and then rose in the 1950s. Chart 2 shows the annual and annualized decade returns of the 10‐year U.S. Treasury for the 1940s, 1950s and 1960s.  

Chart 2Source: Morningstar Direct

What this period meant for Bond investors was an annualized return of 1.65% for the first 20 years and 1.9% annualized return for the 30‐year period. Now it is true one could have gotten higher returns in corporate bonds, but not much higher.  

We are living this, at least the 1940s part, today. Chart 3 shows the annualized return of the 10‐year U.S. Treasury over the last 9‐year period. The 10‐year yield on 12/31/08 was 2.25% and closed 2017 at 2.40%. The last 9 years have seen the exact return from bonds that investors achieved in the 1940s.

Chart 3Source: Morningstar Direct

RIAs have relied on fixed income to produce income and reduce the volatility of one’s total portfolio. Today fixed income continues to provide both benefits. The problem is the income level does not meet most investors’ needs for either income or implied total return. Almost all the clients of our RIA firms cannot achieve their investment goals with the current and implied future returns from fixed income.  

To solve their problem, many investors have included High Yield as part of their fixed income portfolio, and to date it has worked well. The 10 Year Treasury annualized at 2.5% over the last 9 years, High Yield has averaged 12.7%. Now it is worth noting that over half of that return came in 2009. And for those who bought the lower rated parts of high yield, below B rated bonds, they were rewarded at even a higher level.  

We have two problems with this solution. First, High Yield securities are not safe, fixed income investments. Now technically they are fixed income, but the asset class with the highest correlation to High Yield is domestic small cap equities, so historical volatility in this asset class is much higher than investment grade.  

Our second problem with High Yield deals with our definition of risk. We believe risk is not just the standard deviation of an asset class or its maximum draw down. We believe the prudent definition of risk lies in an analysis of what could have happened but did not. For example, the price of oil fell from $101.87 to $34.37 from July 2, 2014 to January 20, 2016. Oil then rose in 2016 to $56.85.   What would have happened if oil stopped at $35 and did not rise? Many high yield investments would have been devastated. So, we do not know if investors dodged a bullet, but we do know that no one would feel as comfortable with high yield today if oil had not rebounded.  

There is one more observation on High Yield. While the 10‐year compounded at 1.6% over the last five years, High Yield returned 9.8%. Spreads today are narrow, and this party is over.

We think this is the biggest investment issue investors face today. One cannot rely on fixed income as we have in the past. Projected returns of 2.5% or slightly more will not solve clients’ investments return needs. Investors need to have less, if any, fixed income in their portfolios.  

So, If Not Bonds, What?  

We have written a paper on this topic and would be happy to make it available to you. The answer is what we call Alternative Investments. These are diversified ETF portfolios that are focused on producing portfolios with varying diversification outcomes. These managers have two portfolios, one whose goal is a 5% standard deviation and one whose goal is 7.5% standard deviation. We have two managers we use, Windhaven Investment Management and 3EDGE Asset Management. We have used these firms for over a decade and have seen the benefit of diversifying into this asset class.

The Conservative Portfolio  

Many RIAs set aside 10% to 20% of client assets in a laddered bond portfolio with bonds maturing in two to four years. This is the “Safe Money” portfolio. Today, a laddered investment grade portfolio with maturities equally spread between two to four years would yield 1.8%. So that solution is not realistic for most clients. It is safe, but the returns simply do not solve the investment needs of almost all clients.  

The conservative portfolio of our Alternative Investment managers provides an interesting solution for this “Safe Money” dilemma. These portfolios will have slightly higher levels of volatility than a bond ladder, but they have a chance to achieve a return that solves clients’ return needs. We should also note that this Conservative Portfolio was down 5% in 2008. More volatile than bonds, yes, but it gives your clients a chance to achieve their return needs.  

The Growth or Total Return Portfolio  

We have been using Alternative Investments in clients’ portfolios since 2000. Over this 18‐year period our clients have achieved a 7.3% return with a 7.4% standard deviation. These portfolios consisted of Fund of Funds in the early years and then liquid ETF portfolios for the majority of this time. One other critical statistic is these returns were achieved while getting 45% of the up‐capture of the S&P 500 and 28% of the downcapture. The biggest stress test came in 2008 when Windhaven was down 14.7%.  

These portfolios run by Windhaven and 3EDGE are our recommendation today. They are intended to provide a mix of return and volatility, higher volatility than bonds and lower than stocks, combined with a very good chance of reasonable returns even in a low GDP growth environment.  

Combining Alternatives with a Stock/Bond Portfolio    

The obligation of a fully diversified portfolio is to deliver returns to the client that will meet their long‐term investment needs at a level of volatility that clients can withstand. The problem RIAs face today is helping their clients through what is likely to be a long time when bond investments will not furnish sufficient returns to meet the first part of this objective. Bonds will not provide ample returns to meet almost all of your clients’ investment needs. This is the dilemma that we all face and the reason that all clients need alternatives.  

The historical standard deviation of stocks is 15% and bonds is 5%. We start with the premise that most clients should cut their bond exposure by 50%. If we use just stocks and bonds the 60/40 clients becomes an 80/20 client. This does not work because of the increase in volatility when going from 60/40 to 80/20.  

If we use 15% as the standard deviation of stock portfolios, 5% for bonds and 8% for alternatives, you can see below the impact on total volatility of portfolios.  

A 60/40 stock‐bond portfolio equals a 11% standard deviation assuming no correlation risk.

A 50/30/20 stock‐alternative‐bonds equals a 10.9% standard deviation assuming no correlation risk.  

The discussion on correlation is a long one, and we will leave that for a time we can discuss that at length. Adding alternatives definitely adds some theoretical correlation risk. That said, we would recommend the 60/40 portfolio become the 50/30/20 portfolio. It will give the 60/40 client the combination of potential return and level of volatility they can withstand to put them in a position to accomplish their return goals. 

10 Years from Now  

There will be surprises on the plus and minus sides of investing over the next 10 years; that is a certainty. Low bond returns also look to us like a certainty, while stock market returns look a little cloudier. We will have, though, our correction.  

Getting volatility right for clients has been the job of RIAs over the last 25 years or longer. 25 years ago, we had good projected returns from bonds and we expected stocks to be volatile. We think the next 10 years will be about getting the return portion of risk allocation right.

Ten years from now as we look back to 2018, bond returns will have mirrored the 1940s or a combination of the 1940s and 1950s with returns in the 2%‐3% range. Some investors will stick with a traditional mix like the 60/40 mix of stocks and bonds. We think these people will be very disappointed in their returns. They may get the volatility right, but as clients retire many will not have sufficient capital to live the lives they desire. This is what we see as the big risk of the next 10 years.