FXIS Market Insights

Weekly Snapshot
• U.S. consumer sentiment index at 70.8, up 1.9 points from May
• U.S. personal income in May rose higher than expected to 1.4%
• US savings rate at 6.9% - highest 15 years
• European Central Bank injects liquidity of €442bn in form of a 12-month repo
• The Fed leaves rates and bond purchase program unchanged and suggested inflation remains benign
• FOMC: "the pace of economic contraction is slowing" and "conditions in financial markets have generally improved"
• Russia stocks down more than 20% from its 2009 peak on June 1st
• U.S. final Q1 GDP growth revised to -5.5% annualized
• U.S. Durable goods orders surprisingly rose 1.8% in May
• U.S. New home sales drop 0.6% in May
• Insiders sell shares at fastest pace in 2 years
• World Bank cuts 2009 global growth forecast, says world economy to shrink by 2.9%

Interesting Read
Please take a look at: http://globaleconomicanalysis.blogspot.com/2009/06/long-term-buy-and-hold-is-still-bad.html 

This is a good eye opener and an excellent starting point for some further examination, particularly when it comes to your own investment planning.

First off, let's put a few things into perspective:

When brokers, investment advisors (myself included) and increasingly bloggers point out the obvious by looking at the past via historic charts, tables and other statistics, take it with a grain of salt and verify the information, particularly as it pertains to your own investment objectives. Always remember that the past and past performance cannot guarantee future performance. It's somewhat like driving ahead by looking in the back view mirror...

Having said that, and without undermining the important findings in this article, let's do our own analysis...

The S&P 500 is indeed down over 30% from 10 years ago. On June 25 1999, the S&P500 traded at 1,315.31 and closed this Friday at 918.90, a return of -30.14%. As Mike Shedlock pointed out in his comments, the calculations should include dividends, the addition of which makes this poor performance a little less daunting. If we were to include dividends, the returns would be about -18%, still bad but not nearly as frustrating.

The easiest way for you to verify these calculations is to go to Yahoo Finance and examine ticker symbol SPY (I'm picking an ETF here because it captures the market just like the index but trades like a stock and therefore has the implied dividend calculations in their historic tables). Go to http://finance.yahoo.com/q/hp?s=SPY  and select the reference date to see the prices you wish to examine. The difference of implied sell versus implied buying price would be your hypothetical return. By the way, the same can be done with any individual stock that may be part of your portfolio.

Going one step further, take the first price data available from SPY. On the first day of trading on 29-Jan-1993, SPY closed at $43.94. If you were wise enough to buy and hold the ETF since then, you would have made a handsome 109% return, or about 6.6% annualized. Better yet, include the dividends into the calculation by using the adjusted closing price of 32.95, and your returns jump to 178.73%, or 10.89% annualized. Of course, very few of us invest a lump-sum amount at one point and then hold it forever; a typical investor is funding his investments on regular or irregular intervals over a specific investing period. Therefore, the typical slogan you hear from fund managers and brokers "If you had invested $1000 in 19XX, you could have made Y" is purely academic and does not help much in terms of a true assessment of investment returns.

Instead, consider something a little closer to reality. The following may be a bit of a stretch because we need to include a number of assumptions, but bear with me here, I think it will be worth your while.

Let's say you had invested once a year, by buying the market index e.g. SPY, and continued to accumulate each year and hold the same until today. Add one more parameter and choose your investment time randomly by selecting any given trading date in each investment year until today. Finally, examine whether a fixed $ amount (a.k.a. Dollar cost-averaging) or a fixed number of shares should be purchased. Thanks to our handy helpers (computer & Excel), we can easily plot some hypothetical returns and retest the returns by continuing to randomize each trading date per year. We went one step further still and included commissions of $10 per trade in these hypotheticals; it's not as if we can trade for free as some analytics may suggest. The results may surprise you...


How to interpret this chart
Essentially, we picked one random date each year, assumed the adjusted closing price as our purchase price and calculated the returns of each of these purchases in every year since 1993 if held until June 25, 2009. We then repeated the same process with other randomly chosen dates in each of the previous years; in fact, Excel did this 100 times and the above chart shows the plotted annualized returns for each iteration. Using the $ cost-averaging method, you would have achieved an average annualized return of 1.05%. Buying a fixed number of shares would have resulted in substantially lower returns of only 0.17% based on our sample tests. For those of who you are interested, the standard deviation of the hypothetical returns from $ cost-averaging was 0.13% and about the same (0.12%) from the fixed share trading method. You can also visually grasp that the dispersion of the plotted data isn't too far from the mean i.e. the volatility of the returns is low indicating that results are within a narrow range despite the random parameters chosen. Doing this entire process many times over and generating thousands of random samples, we found that the returns are always averaging around 1% for the $ cost-averaging method and close to 0% for the fixed share method. Again, please keep in mind that we are looking backwards from today...

The results kind of stink in the face of the notion of timing the market. By the same token, any randomly chosen date in 1993 for a lump-sum investment held until today would have averaged about 10% annualized.

Now let's assume that buy some magical feat the US market was extremely bullish for the rest of the year and the SPY would end the year at $125, about 20% below the all time high in 2007. Doing the same hypothetical random share purchases again, we are now looking at much better results.


Using the $ cost-averaging method, you would now achieve an average annualized return of 3.27% as opposed to the fixed share method with 2.09%. Both strategies would be better but are still lagging behind the the one-off lump-sum purchase which would have returned over 15% annualized.

What does this mean?
For one thing, the more you trade, the higher the probability that you leave some returns on the table, that is if one believes that on average the US economy and hence the US equity markets are trending higher in the long term (i.e. 20 years time horizon); typically, a single lump-sum purchase would out-perform any regular contribution method, however randomly chosen.

If you are an active trader, timing is everything and if timed right, you can greatly outperform the various buy and hold strategies (in theory). But can you really time the markets?

Having done our random purchases and plotting the returns so nearly dispersed to the mean, we have a very strong case that timing for a buy and hold strategy over the long term was basically irrelevant. It would be more difficult to prove that timing the market by buying at the right time and taking profits at again the right time would achieve superior returns. But that's exactly the million dollar question: When do you know that it's the right time? What method, tool or analytical method can do that successfully over a long period of time?

To be perfectly honest, I don't know of any and the fact that over 70% of mutual fund managers don't even outperform the market, let alone be consistently profitable over a 20-30 time horizon, leads me to believe that very few can. So the question to those who say "buy and hold is dead" would be: What else can we do that achieves better returns than the market?

CD laddering as described in this article may be a good alternative, but again this too very much depends on timing. At current rates, a 6-month CD pays somewhere between 0.5%-0.7% depending on bank and issuer, hardly a rate at which an average return of 78% in 15years is possible. My memory might fail me here, but I don't recall extended periods of time in the past 15 years when a 6 month CD paid more than 4% (an average rate of 3.9% compounded would have been required to achieve the 78% compounded return). But the concept is good and could possibly be achieved with longer maturities, say 1-2 year CDs. Further, CDs are FDIC insured, currently up to $250,000. For investors closer to retirement, this appears to be an ideal strategy to gain secure and regular income. Personally, I would prefer this anytime over a Bond portfolio which is typically a lot more sensitive to swings in interest rates.

For those of us who are not quite there yet (retirement that is) and given the projected state of future government finances perhaps never will, an investment portfolio made up entirely of CDs may not be that appealing, certainly not at current rates. But going back to the title of this article, an "all equity" long-term buy and hold is definitely bad advice, always has been in my point of view. A much more sensible and equally easy to implement alternative is to take another one of Jack Bogle's investment mantras: Take your age and use that as a percentage guidance of the fixed income (Bonds, CDs, Savings) component within your investment portfolio (see Market Insights 10-Oct-2008). In other words, if you are 50 years old, half of your investments should be in Bonds or preferably CDs, the remainder in low cost Index Funds. With a simple portfolio allocation such as this one, you are theoretically guaranteed to always outperform the equity market during in bad times. However, you may also give up some returns when equities are soaring - just keep in mind you can't have it both ways...

I realize that these findings may have raised a few eye brows but I'm more than happy to accept any questions or comments - email:  

Continue reading and get a free copy of our weekly newsletter here...

Good luck & good trading!