Saturday, January 26, 2013

Will investors ride bull market?

My bullish call on stocks at the end of 2012, so far so good:

http://www.csmonitor.com/Business/2013/0126/Will-investors-ride-bull-market

This wasn't about predicting earnings, the jobless rate, etc.  Simply the market had been in a downtrend right after the election, it then began an uptrend in mid November.  Once stocks start an uptrend it is highly likely that the uptrend will last for a while.  Simple as that.

The Risk of Safety

Yet another article about the bubble in the bond market:

http://online.wsj.com/article/SB10001424127887323854904578261733526191140.html?mod=ITP_businessandfinance_5


Most investors today have never seen a bear market in bonds, interest rates have been steadily decreasing since the early 80's so investors have come to look at bonds as "safe" and stocks as "risky".  True safety is being in harmony with market trends, in stocks and bonds, investing in an asset class when the rewards outweigh the risks and not investing in it when the risks outweigh the rewards.  With interest rates so low at best you are looking at earning just the coupon on bonds, at worst you are looking at capital losses when interest rates increase.   Not much reward there and a lot of risk, that is not my definition of "safety".

Wednesday, January 23, 2013

Machine Learning in Financial Trading

http://marketsmedia.com/machine-learning-in-finance-theory-and-applications/?email=feat

Just read an interesting article in Markets Media about machine learning in trading.  Our chief systems designer, Murray Ruggiero, wrote a book about this a while back regarding using neural networks---computer programs that learn and make predictions-- to trade.  To me, the most interestsing use of this technology is when to use trend following strategies and when to use mean reversion strategies.   Trend following strategies work best in trending markets while mean reversion strategies work best in choppy markets.  It is easy to figure out what type of market we have been in, but to be able to accurately forecast what type of market is likely in the future would allow the trader to find an optimal allocation between trend and mean reversion strategies. 

Friday, January 18, 2013

Some Common Sense on the 4% Rule



The 4% withdrawal rate in retirement has become the gold standard in financial planning.  Unfortunately, while it looks good in practice it might not work in real life (meaning the client could be at great risk of running out of money).  The logic beyond the 4% rule is that in retirement a client could withdraw 4% of their portfolio every year, for example a $1mm portfolio could sustain withdrawals of $40,000/year over the retirement period.  The problem with this is that you need a certain growth rate in the market and you need to avoid large losses to make this work and unfortunately the market doesn't really care what you need, it will do what it will do. 

The article above in FA  Magazine not only tackles this but also suggests one answer is to use tactical portfolios to avoid large drawdowns.

http://www.fa-mag.com/news/the-4--rule--static-decisions-in-a-dynamic-world-13055.html

Wednesday, January 16, 2013

Simple Sector ETF Momentum Strategy

Just read a post on CXO Advisory (full post available to subscribers only):

http://www.cxoadvisory.com/4400/momentum-investing/simple-sector-etf-momentum-strategy-performance/#more-4400

Which tested a number of simple strategies to switch among different sector ETFs:

Materials Select Sector SPDR (XLB)
Energy Select Sector SPDR (XLE)
Financial Select Sector SPDR (XLF)
Industrial Select Sector SPDR (XLI)
Technology Select Sector SPDR (XLK)
Consumer Staples Select Sector SPDR (XLP)
Utilities Select Sector SPDR (XLU)
Health Care Select Sector SPDR (XLV)
Consumer Discretionary Select SPDR (XLY)

What they found echoes our research.  It is relatively easy to create a sector momentum strategy that beats the market but not by enough to really make it compelling, especially since there are many other momentum based strategies that do much better than sector momentum.  Here is their summary:

In summary, evidence from a limited sample period suggests that a simple sector ETF momentum strategy fares well compared to the overall stock market and adds some value to a simple (equal-weighted) diversification approach, but this added value may be dissipating.

 


Friday, January 11, 2013

The TTM Approach to Tactical Asset Allocation


 
Traditional asset allocation theory relies on a number of key tenets:

·         Markets can go down in value but they always come back.  Therefore, long term investors should not be concerned with market downturns.

·         Different asset classes (large stocks, small stocks, international stocks, growth, value, etc) are uncorrelated.  Therefore, spreading your portfolio among these different asset classes can produce higher returns with less risk.

·         Markets are unpredictable.  Therefore a passive approach is best.

Markets Don't Always Come Back Over The Investor's Time Horizon

Unfortunately, as many investors have discovered the hard way, these tenants do not hold up in real life.  Looking back over hundreds of years of data, whenever markets go down they do eventually come back.  However, that may not be over the investors time horizon.  Studies showing how markets react over time typically use periods of 70 years or more.  That is hardly relevant to most investors.  During such a long time there will be some ups, some downs, and some sideways movements.  The average investor who retires at age 65 might have a time horizon around 20 years.    If we look at the market over 10 and 20 year blocks the story is very different.   For example, from 1965 to 1982 an investor who bought and held the market would have made 5% total.  If he started with $100, he would have had $105 after 16 years.  More recently, if an investor retired on December 31, 1999 with $1mm and put it into the S&P 500, at the end of 10 years he would have $900,000 left.  If he had assumed a 10% return in retirement over a 20 year period then he would have to average 22%/year over the next 10 years just to get back on track.    If we assume he bought into the idea that he could withdraw 4% in retirement ($40,000/year) then the picture is even worse---his account would have shrunk to $484,000.  He is now at great risk for running out of money.  Over the same time period a $1mm investment in the NASDAQ on December 31, 1999 would have been worth $557,000 after 10 years.   

Below is a chart of the Bear Markets in the S&P 500 since 1929, their duration, the amount of decline, and how long it took investors to get back to even.

S&P 500 Bear Markets
Duration
% Decline
Time to Breakeven
Sept ’29 – Jun ’32
33 months
-86.7
25.2 years
Jul ’33 – Mar ’35
20 months
-33.9
2.3 years
Mar ’37 – Mar ’38
12 months
-54.5
8.8 years
Nov ’38 – Apr ’42
41 months
-45.8
6.4 years
May ’46 – Mar ’48
22 months
-28.1
4.1 years
Aug ’56 – Oct ’57
14 months
-21.6
2.1 years
Dec ’61 – Jun ’62
6 months
-28.0
1.8 years
Feb ’66 – Oct ’66
8 months
-22.2
1.4 years
Nov ’68 – May ’70
18 months
-36.1
3.3 years
Jan ’73 – Oct ’74
21 months
-48.2
7.6 years
Nov ’80 – Aug ’82
21 months
-27.1
2.1 years
Aug ’87 – Dec ’87
4 months
-33.5
1.9 years
Jul ’90 – Oct ’90
3 months
-19.9
0.6 years
Mar ’00 – Oct ’02
31 months
-49.2
4.7 years
Oct ’07 – Oct ’08
17months
-56.8
Unknown

Source: Telephone Switch Newsletter, Summer 1992. Updated by the National Association of Active Investment Managers, Inc. through 2012.

One thing that this chart makes clear is that Bear markets are fairly common, very painful, and it can take a long time for investors to break even.  Between 1929 and 2008, there have been 15 bear markets with an average decline of 39%. During those 81 years, a new bear market began on the average of every five years, and lasted 18 months. After the bear market bottomed, omitting the 1929 crash, it took an average of 3.6 years just to break even.

Bear markets have a number of other consequences. 

·         While markets eventually come back, investors lose time and opportunity during them.  According to the National Association of Active Investment Managers over the past 70 years, the major indices spent nearly 60% of the time sitting out bear markets and then returning to earlier highs. Only about 40% of the time were real gains being made. 

·         Investors rarely buy and hold.  Instead, they tend to sell out when the pain of loss becomes unbearable, typically at, or near, market lows.  Then they tend to buy back in when the pain of watching everyone else make money becomes unbearable, typically at, or near, market highs.  According to Dalbar Financial Services for the period 1984-2003, the average return of the S&P 500 Index was 12.2%. Equity mutual fund investors during the period achieved an average return of 2.67%.  This is due to the average investor selling at lows and buying at highs.

Correlation

Asset classes that may be uncorrelated during up markets tend to become correlated during down markets as investors sell everything in a panic.  Below is a chart of the returns of the S&P 500, international developed stocks, emerging market stocks, large cap growth stocks, large cap value stocks, and small cap stocks in 2002 and 2008:

Category
2002 Return
2008 Return
S&P 500
-22.10%
-37.00%
MSCI EAFE (Foreign Stocks)
-15.94%
-43.38%
Morningstar Emerging Markets
-6.05%
-54.44%
Morningstar Large Cap Value
-17.21%
-35.85%
Morningstar Large Cap Growth
-27.83%
-41.47%
Russell 2000 (Small Cap)
-20.51%
-33.64%
Source: Morningstar.com

 

A "diversified" portfolio of large stocks, small stocks, international stocks, emerging market stocks, growth, and value, would have offered no protection from bear market declines.

Markets Are Not Predictable But True Trend Changes Are Recognizable

Nobody can predict markets however there are a number of factors that investors can consider to put the odds in their favor.   There are five outcomes for investors in the stock market over any week, month, year, etc:

1.       You can make a lot of money

2.       You can make a little money

3.       You can be flat---not make any money, not lose any money

4.       You can lose a little money

5.       You can lose a lot of money

Fluctuations happen, so there is no way to avoid small losses.  However, large losses of the type that happen in Bear markets can be avoided.  True trend changes don't happen overnight, they take time to form.  While nobody can accurately predict Bear markets it is relatively simple to see a true trend change.  A change in trend may not persist and turn into a Bear market but it substantially increases the risk of large losses and limits the potential rewards. 

Putting the Odds In Your Favor

There are typically three types of markets:

1. Markets that go up in a fairly straight line (Bull Markets)

2. Markets that go down in a fairly straight line (Bear Markets)

3. Choppy markets that could go in either direction

In a traditional asset allocation framework the investor would treat each market the same.  He would ride the bull market up, he would ride the bear market down, and he would move up and down with the choppy market and hope that he ended up positive.   Using Tactical Asset Allocation (TAA), each type of market should be handled differently.   In Bull Markets, investors should be mostly, or completely in stocks.  In Bear Markets, investors should be mostly, or completely out of stocks.  In choppy markets investors should use a mean reverting or counter trend approach to make money from market overreactions.  This is where having a model based approach is key as it removes the emotion from the equation.  When stocks weaken then a momentum or trend following model based approach would move out of stocks into bonds and/or cash.  When stocks strengthen then a momentum or trend following model based approach would move back into stocks. 

How Markets Work

Markets are not efficient because they are traded by people.  People bring their emotions to their trading---greed, fear, and ignorance.  This creates a number of opportunities for profitable trades, including:

1. In the intermediate term assets that have gone up tend to continue to go up.  Basically, whatever is hot, stays hot for a long enough time to profit from it. 

2. In the short term markets overreact on the upside and downside and eventually snap back. 

At TTM we take advantage of both of these opportunities using momentum and counter trend analysis.

The TTM Approach

We use an objective allocation model that analyzes two categories of information:  momentum and counter-trends.   Momentum measures tell us the strength of stocks vs. bonds and which stock and bond sectors are the strongest.  Counter-trend analysis identifies possible short term market turning points and where the market may be overbought or oversold.  Momentum measures work best in straight up or down markets as in those environments a move down or up in stocks or bonds tends to be persistent.  Counter trend analysis works best in choppy, directionless markets. 

Momentum

Momentum is the documented tendency of investments to persist in their performance. In other

words: stock, bond, commodity, or currency sectors that outperform during a given time period

tend to continue to outperform. More than 300 academic papers have been published illustrating

that momentum outperformance exists in just about every asset class.  We use momentum to decide how to allocate strategies among stocks and bonds and then to decide which stock sectors and/or bond sectors to include.

Counter Trend Analysis

Momentum based models work very well in straight up or down markets but tend to be much less effective in directionless markets.  They also will typically give some of their profits back at market turning points.  counter-trend analysis is equally as effective as momentum analysis but completely opposite in methodology.

 

Counter-trend trades are typically much shorter in duration than momentum trades and usually have a much lower win/loss ratio (winning trades are smaller and losing trades are smaller).   Counter-Trend models look to buy into oversold markets, expecting them to bounce back, and sell into overbought markets, expecting them to selloff.  Over the short term, markets are dominated by noise, fear, and greed.    This causes short term dislocations and deviations where market prices stray too high or too low and then snap back. 

Counter trend strategies can thrive in directionless markets and react quickly to major turning points.  They can be less effective during steady trending markets. 

 

 

Achieving Market Beating Returns with TAA

Most TAA strategies have the stated goal of beating the market over a typical market cycle (4-6 years on average).  They try to do this through a number of different strategies:

·         Apply a rotation strategy, sector, country, and/or style.  These managers will try to outperform by being in the right style, sector, and/or country.  Based on our research this is hard to pull off on a consistent basis and runs the risk of a situation where the market goes up but the sectors, styles, and/or sectors the manager is in do not. 

·         Create a portfolio of non-correlated asset classes.  Such a portfolio might have tactical allocations to US Stocks, Bonds, REITs, International Stocks, and Commodities.  This is also hard to pull off as by nature these asset classes tend to be uncorrelated and could drag on portfolio performance if one area is in a drawdown.

·         Buy individual stocks. This can work if the manager can consistently pick winners but this also exposes the investor to company specific risk (ie. AIG, Bear Stearns, Lehman, etc).

At TTM we take a multi-factor approach to try to beat the market:

1. Being heavily invested in stocks when our momentum models are positive

2. Being in the strongest market cap sector among large cap stocks, mid cap stocks, and small stocks.  Our momentum models determine which market cap sector has the strongest momentum.

3. Using counter-trend analysis to buy into short term lows and sell into short term highs.  This allows us to profit from short term market turning points and choppy markets.

4. Moving out of stocks and into bonds during times of market turmoil, protecting gains and avoiding losses.

5. Applying a prudent leverage factor during sustained uptrends.

How We Use Leverage

Leverage can sometimes have a negative connotation for investors and rightfully so.  Used inappropriately, leverage can increase portfolio volatility and magnify losses.  There are times when the risk of being invested in the market outweighs the rewards and there are times when the rewards outweigh the risks.   During times when market risk is low and potential rewards are high(during sustained market uptrends and when the market is oversold over the short term) then the use of leverage is warranted and can magnify returns without taking un-necessary risks.  During times when market risk is high and potential rewards are low (during sustained market downtrends and when the market is overbought over the short term) then the use of leverage is not warranted as it would increase risk and potential losses.

How We Hedge Market Risk

How we hedge against market risk during market downturns is more important than how we try to beat the market during upturns.  There are a number of ways tactical managers can use to hedge portfolios----moving to cash, moving to bonds, and/or using inverse funds.  We have chosen not to use inverse funds, we like making money when the market goes down but we are unwilling to take a chance of losing money when the market goes up.  We have developed a number of profitable tactical methodologies that use inverse funds but they are only right a little over 50% of the time, that isn't good enough for us.   Instead, we prefer to move to bonds if they are attractive, and cash if they are not.  For our bond allocation we have chosen long term Treasury Bonds and the Barclays Aggregate Bond Index as our preferred hedges.  While Treasuries are typically a great hedge to the stock market, they can also be quite volatile, so we have exposure to them but it is limited.  The Barclays Aggregate Bond Index gives us the ability to earn more than cash without the volatility in Treasuries.

Putting It All Together

It is clear that traditional asset allocation doesn't work.  Market declines happen too often, last a long time, and take too long to recover from.  Our approach offers the ability to profit in market uptrends while avoiding multi-month market declines.

How We Use Bonds


Typically, investors hold bonds in their portfolio to mitigate the volatility of stocks.  While bonds may reduce the portfolio volatility and drawdowns of a stock portfolio somewhat, it would take an extremely large bond allocation, along with an extremely large reduction in potential portfolio returns, to have any meaningful impact.   

Reducing Portfolio Volatility and Drawdowns--The Traditional Approach

Very few investors can tolerate the volatility and drawdowns of a portfolio that is 100% weighted towards stocks.  From October 2007  to March 2009 a portfolio invested solely in the S&P 500 would have endured a 60% drawdown.  Not many people could endure that kind of loss without selling in a panic.  Bonds are often not correlated with stocks, especially during market downturns, and they typically are not subject to the same magnitude of drawdowns that stocks are (depending on the type of bond of course).    In this case, adding bonds to a portfolio would likely reduce volatility while also reducing returns.  For example, in 2008 the S&P 500 Index dropped 37% while the Barclays Aggregate Bond Index ETF (AGG) increased 7.57%.  An investor who had a portfolio that was 60% in the S&P 500 and 40% in the AGG would have lost about 20%.  This is better than losing 37% but it is still an unacceptable loss.  During up years for the market this same investor is likely to give up a lot of market upside for not much protection on the downside. 

Investors will often think of bonds as risk free or near risk free.  Bonds have been in a bull market (caused by a steady decline in interest rates) since 1982, this covers the entire investing experience of most investors today.  If they have never seen a bear market in bonds they start to believe that it can't happen.  Bond prices move inversely with interest rates, if interest rates rise then bond prices will go down.  Nobody can predict whether interest rates will rise in the future but common sense suggests that they will.  The Federal Reserve has been artificially keeping rates low since the 2008 crisis, this will not last forever.   Traders also make decisions on whether to allocate money to stocks vs. bonds based on risk and return.  In times of high interest rates and/or financial crisis, traders move money to bonds.  In times of low interest rates and rising stock prices, money flows to stocks.  As the stock market continues to recover from the 2008 crisis, more and more money will flow out of bonds, causing prices to go down, and into stocks. 

Bonds can also decline in value for other reasons besides interest rates.  Higher yielding areas of the bond market, such as high yield and emerging market bonds, had drawdowns of 25%+ during 2008. 

Reducing Portfolio Volatility and Drawdowns--The TTM Approach

Instead of having a fixed allocation to bonds that could-----decrease portfolio drawdowns somewhat, decrease portfolio returns, and lose value when/if interest rates rise----we maintain a tactical allocation.   When our momentum measures show that bonds are stronger than stocks this indicates that the risk of a decline in stocks is high and the potential rewards are low.  In this instance we can shift all the way to 100% in bonds.  Not only would this enable us to reduce drawdowns substantially during bear markets, it also allows us to potentially make money in a market downturn.  When momentum shifts back towards stocks it indicates that the potential rewards in the stock market are high and the risks of a decline is not as high.  In this case we can move out of bonds and go to 100% stocks, avoiding the drag on portfolio returns that comes from owning bonds. 

If our momentum measures favor bonds over stocks, then a decision must be made about which bond sector(s) we should allocate to.  Typically, we only use bond sectors that are easy to buy and sell through actively traded ETFs---investment grade corporate, high yield corporate, emerging market, Barclays Aggregate Bond Index, Treasury Inflation Protected Securities (TIPs), and/or Treasury Bonds.  During times of great market turmoil, Treasury Bonds tend to be the best bond sector as traders prefer the perceived safety of Treasuries, regardless of interest rates, to the risk of stocks. 

At some point the bull market in bonds will end---for it to continue interest rates need to decline from here which is possible, but not likely.  Once it ends, investors that have fixed allocations to bonds will suffer losses.  Until it ends, it is likely that bonds will drag down portfolio returns as a major part of their return comes from interest rates, which are currently low.  By using bonds tactically to hedge whether they are in a bull or bear market is irrelevant.    Regardless of the level of interest rates bonds are a safe haven at times of turmoil in stocks.  Investors are willing to earn very little, if anything, in bonds to avoid potential losses in stocks. 

 

Monday, January 7, 2013

I Actually Agree With Goldman Sachs

Why Goldman Thinks You Should Dump Bonds Now

Goldman Sachs has issued a warning to their clients to get out of bond funds.  I actually agree with them, however, not for the same reason.

Goldman is predicting that interest rates will increase:

"A reversion of risk premiums to historical averages of 6% nominal rates (3% real rates and 3% inflation) would suggest estimated losses in portfolios with bond durations of 5 years of 25% or more," equity strategist Robert D. Boroujerdi said in a note."
 
Maybe they are right, maybe they are wrong, at some point interest rates will probably increase for a number of reasons:

1. They are real low now and don't have a lot lower they can go
2. A lot of bond buying has been flight to safety, at some point things will improve globally and that money will probably shift to equities.
3. The Fed is keeping them low, they can't do that forever

However, who knows when this will be, that is what makes investment decisions based on predictions so hard. 

The real reason that investors should get out of bonds and into stocks instead should be based on market trends.  Momentum and relative strength cleary favor stocks right now, as long as that is the case being in stocks offers a better risk/reward than bonds.  That doesn't mean stocks will continue to rise from here, it just means that the odds of that happening are in your favor.  When/if that changes investors will be better off in bonds. 

Wednesday, January 2, 2013

The Only Statistics That Matter


There are a bunch of statistics that Wall Street uses to try to help investors evaluate potential investments.  Most of these measures are useless or close to useless.  The two most important statistics that will help an investor determine whether they will be happy with a potential investment are Maximum Drawdown and MAR ratio.

Maximum Drawdown is the maximum peak to trough loss in an investment.  For example, let's say you put $100 into a mutual fund in January.  In August it is worth $200 and in December it is back down to $110.  The mutual fund company will tell you that you made 10%, and they would be right, but it probably won't feel that way.  Instead you may look at it differently thinking you had  $200 in August and $110 in December so you lost 45%.  This is the maximum drawdown.    Now you might be thinking---why do I care about Maximum Drawdown, I made 10% I would be happy with that?   That's fine, assuming you invested in January, but what if you invested $200 in August?  The mutual fund would be up 10% for the year but you would be down 45%.  Drawdowns occur in just about every investment, it is just a matter of when.    They don't matter as much if you have held the investment for a while and have large gains, but they matter a lot more if they happen when you start off in an investment.    Of course past returns don't predict future results but if an investment has had a 20% drawdown in the past it is likely that it would have another similar drawdown in the future.

Knowing Maximum Drawdown is half the battle,  any investment that has potential to appreciate will have a drawdown, so how much is acceptable?   Part of the answer to that depends on your risk tolerance and what the money is for but the main question you should have is are you being compensated for the risk you are taking?  This is where MAR ratio comes in.  MAR is the average annual return of an investment divided by Maximum drawdown.  When we evaluate investments or strategies we don't look at anything with a MAR under 1, and much prefer MARs approaching 2.   Using an S&P 500 Index ETF (SPDR S&P 500 symbol SPY) as an example, according to Morningstar the average annual return over the past 10 years is 7%, we also know that the S&P 500 had a drawdown of about 60% from October 2007 to March 2009, so the MAR for this fund would be .11 (7/60).  For us to consider an investment that had a 60% drawdown we would need an average annual return of at least 60%. 

Buy and hold stock based investments and investment strategies will always have large drawdowns as they do nothing to stem losses when the markets go down so it would be nearly impossible to find a buy and hold stock based strategy with a MAR ratio anywhere near 1.  Trend following and  other tactical strategies are the only way to find higher MAR ratios.  Trend following strategies will still have drawdowns, but they tend to exit markets before bad turns into really bad.