Saturday, 4 April 2015

The Market and useful history notes coagulated

100 Years of Stock Market History (log graph thru 2012)

In times of turmoil, such as a financial crisis, I look for one of those big charts with an arrow that says, “You are here.” It is in that spirit that I offer the following long-term log graph summarizing over 100 years of DJIA (Dow Jones Industrial Average) performance / history. I will defer most of my analysis until later, and for this post rely mainly on what one of my statistics professors used to call “interocular trauma.”

Dow Jones 100-Year Stock Market History Chart

stock market history chart year-end 2012 (Dow Index closing prices since 1900 log graph)
Dow Index 100-Year History Chart

Stock Market Performance Since 1900 Has Alternated Between Excitement and Disinterest

Above is a graph of stock market (Dow Jones) performance since 1900 (click on image to enlarge it). It shows year-end closing prices through 2012. (See Yearly Returns for a bar chart of the returns each year.) While some describe this history as
a steady long-term upward trend, to me it appears to show alternating periods of excitement and disinterest. For example, the periods from ’33 to ’65, and from ’82 to ’99 were periods of excitement. From ’33 to ’65 the average return was about 7% per year, plus dividends -- for a total of approximately 10%. From ’82 to ’99 the average return was about 15% per year, again plus dividends – though dividends in recent decades were significantly smaller than they were in earlier decades.

The Long Flat Periods

On the other hand, the 1905 close of 96 was not permanently eclipsed until 28 years later -- 1933; the 1965 close of 969 was not permanently eclipsed until 17 years later -- 1982. I use the word “disinterest” to characterize these long flat periods. (Note: This is a log graph. If you are not familiar with them, see About Stock Market Log Graphs.)

In the long term, you would expect that stock market performance should approximate the performance of the underlying businesses. Therefore, an obvious interpretation of the chart is that the stock market periodically gets ahead of itself by increasing faster than the underlying businesses, and then has to wait for the “real” value of the underlying businesses to catch up during the long, flat periods of “disinterest.” If that is the case, we could well be in another one of those periods of “disinterest” -- though when you’re actually in one of those periods, you may find other words more descriptive….

Note: The above chart and discussion ignores the impact of inflation. To see the long flat periods adjusted for inflation, see 100 Years of Inflation-Adjusted Stock Market History. Warning: not for the faint of heart!

The Monthly Update, & Adding the 25-Year Moving Average as a Support Level

The December 2012 Stock Market Performance post includes a recap of the most recent month and year-to-date, plus comparisons to important milestones such as all-time highs and crash lows. In addition, it includes the most recent projection of 10-year market returns. The 25-year moving average can be a useful addition to the above graph. As discussed in Dow 25-Year Moving Average History, the market has very rarely fallen below its 25-year moving average. That is, historically this moving average has been a reliable support level during secular bear markets. That graph is updated infrequently, as appropriate.


Average Stock Market (Dow) Returns

The average return starts at 11.8% for the one-year returns, drops rather sharply to 10.6% for the two-year periods, and then declines gradually to 9.9% for the 100-year periods. Historically, the longer the holding period the less likely it was to end with the investor losing money, and, the closer the return has been to the long-term average of around 10%. In fact, the maximum, minimum and average returns are all converging toward the long-term average return.

Based on this history, it appears that readers who are planning to buy and hold for 100 years or more are "assured" of performance very close to the long-term average return of around 10% per year -- regardless of when they buy. For those of us holding for shorter periods, there is a significant difference between the minimum and maximum annual rates of return that gets smaller the longer we hold.

Checking through history ,
As they happen every 4.4 years on average.


Statistical response to a DOWN January

A down January is a bad omen for the stock market. Yale Hirsch of the The Stock Traders Almanac suggests that since 1950, every down January in the S&P500 preceded a new or extended bear market, or in some cases, a flat market. They go on to further suggest that down January’s are followed by substantial declines averaging -13%
The Standard & Poor’s 500 Index has returned 24 percent on average in years it’s risen in both January and February, a bullish sign for 2013, according to S&P.
The S&P 500 climbed in both January and February 26 times since 1945, Sam Stovall, S&P’s New York-based chief equity strategist, wrote in a note. All 26 years ended with positive returns when including dividends, the data show.

The last two years have seen back-to-back gains in the first two months, leading the S&P 500 to a 16 percent advance in 2012 and a 2.1 percent rise in 2011, including dividends. The biggest advance when the index rose in January and February was the 52 percent rally in 1954. 2013 saw a nice 27% rise .

The two "largest" price drops during this bull market came when the "Fed was (getting) out of the market," says Bespoke co-founder Paul Hickey. Both times, when the Fed withdrew stimulus, the market suffered a drop of more than 10% -- the conventional definition of a correction.
In 2010, the S&P 500 fell 10.3% from May 12 to June 7. And in 2011, the benchmark index plunged 17.3% between July 7 and August 8, Bespoke data show.
As investors around the world continue adjusting to the idea of less market support – or "the end of free money" -- from central banks globally, including the Fed, the market is likely to stay volatile, says Ron Florance, managing director of investment strategy at Wells Fargo Wealth Management.
"The Fed's action represents the continuing transition that is occurring in the global economy following the financial downturn and the recovery period that has followed," Florance says. "We expect financial markets to respond with a measure of volatility as the 'normalization' process unfolds."

Should you sell in April and go away?
It’s an odd question, I admit. Widespread talk of selling usually doesn’t begin until late April, when investors each year are reminded of the famous seasonal pattern “sell in May and go away.”
But it’s precisely because it is so well-known that some followers of this seasonal tendency wonder if they should act sooner rather than later. Waiting until May Day runs the risk of selling at the same time that a large number of other investors are doing the same.
Fortunately, we have real-world data on two attempts to get a jump start on the “sell in May and go away” pattern. The first is the “Almanac Investor Newsletter,” edited by Jeffrey Hirsch, and the other is Sy Harding’s “Street Smart Report.”
Both pursue surprisingly similar modifications to this basic seasonal pattern. Each relies on a technical indicator known as MACD to pinpoint the precise day on which they enter and exit the market. (MACD is a short-term momentum indicator, standing for moving average convergence divergence.)
The Hulbert Financial Digest has track records for both market timers’ modifications of this seasonal pattern dating to mid-2002, nearly 13 years ago. The HFD calculates their returns on the assumption that, when they are invested in stocks, they earn the return of the Wilshire 5000 Index; otherwise they are assumed to be invested in 90-day Treasury bills.
Annualized return 6/2002 to 2/2015Risk level (100 = market)Sharpe Ratio
Buying & holding7.7%1000.13
Mechanically selling every 4/30 and reinvesting every 10/317.9%62.10.20
Sy Harding’s modification of “Sell in May and go away”9.2%61.00.24
Almanac Investor’s modification of “Sell in May and go away”8.0%64.50.20
As you can see from the accompanying table, a buy-and-hold strategy since mid-2002 has produced a 7.7% annualized return. Automatically going to cash every May Day and re-entering the market on Halloween would have done slightly better with a lot less risk — which is why it comes out well ahead of buying and holding on a risk-adjusted basis (as indicated by a higher Sharpe Ratio).
Harding’s modification of the Halloween Indicator did even better still, producing a 9.2% annualized return over the same period — 1.5 percentage points per year more than a purely mechanical application of this seasonal pattern, and 1.3 percentage points ahead of a buy-and-hold. Even better, this market-beating return was produced with 39% less risk, which means it’s even further ahead of buy-and-hold on a risk-adjusted basis.
In fact, Harding’s modification of the Halloween Indicator is in fourth place for risk-adjusted performance since mid-2002 out of the 91 stock-market-timing strategies the Hulbert Financial Digest has tracked over this period.
To be sure, Hirsch’s modification of the Halloween Indicator performed less well than Harding’s, producing an 8% annualized return while incurring 35% less risk than the market itself. It still came out ahead of buy-and-hold, however, on both an unadjusted and a risk-adjusted basis.
What about this April? To be informed when those two advisers actually trigger their “sell” signals, you would need to subscribe to their services. But one way in which their MACD-based systems could trigger an early “sell” signal (but not the only way) would be for the market to be strong for a few days and then quickly drop back.
Given the stock market’s impressive rally on Monday, such a “sell” signal could come fairly soon. If that happens, followers of this seasonal pattern might consider going to cash — thereby taking off the rest of the spring and summer from stressing about the market.

Three Losing days begin the year

In an email on Tuesday evening, Jonathan Krinsky at MKM Partners noted that not only does the S&P usually gain ground when it starts the year with three losses, but the index has performed better than its historical average in those years. 
During the eight years since 1928 that the S&P started with three-day losing streaks, the index has returned 8.36% on average. For all years since 1928, the S&P has returned 7.53% on average. 
Via Krinsky, here is what happSince 1928, the Standard & Poor's 500 has started the year with three straight losing days eight times.
ened during each of those eight years:
  • 2014: Started with three straight down days. Year ended +11.39%.
  • 2005: Started with three straight down days. Year ended +3%.
  • 1991: Started with six straight down days. Year ended +26.31%.
  • 1985: Started with three straight down days. Year ended +26.33%.
  • 1978: Started with seven straight down days. Year ended +1.06%.
  • 1977: Started with three straight down days. Year ended -11.50%.
  • 1968: Started with three straight down days. Year ended +7.66%.
  • 1956: Started with three straight down days. Year ended +2.62%.
Last week, we noted another market factoid from Krinsky, which showed that this was the first time since 2008 that the market finished one year and started the next with losing sessions. 
That is something that happened 10 times since 1980, so it's not all that uncommon, and in seven of those 10 years the S&P gained ground. 
Of course, this makes some sense: Stock markets have spent more time going up than going down

On Wednesday, all major stock indexes gained more than 1%, breaking 2015's losing streak.

No comments:

Post a Comment