March 31, 2008

• Q1 20007 was the worst quarter for the market in over five years, since Q4 2002.  The S&P 500 was down 9.92!
• As weak as the economy is, we still do not forecast a technical recession (two quarters of contracting GDP growth) in 2008.  (For the sake of fluidity, we refer to the current economic slowdown as a recession in this update.)
• The recession likely started in December 2007 / January 2008.
• The average stock market slide in the ten recessions since WWII has lasted 5.5 months from the beginning of the recession.  We are working through a bottoming process that likely has less than a few months to complete (April/May).
• The risk is that our S&P 500 risk level of 1,300 points may yet be broken as we approach the Presidential general election.
• Whatever the ultimate low price point for this recession cycle, recession lows have been a table-pounding buying opportunity for the next 3-, 6-, 9-, and 12-months.
• We continue to expect that 2008 will be a tale of two markets where the majority of the weakness occurs in the first four-to-five months of the year followed by a rejuvenated stock market anticipating a typical post-recession business cycle upswing.

We still have not seen two back-to-back quarters of negative economic growth, as measured by Gross Domestic Product (GDP).  That is the traditionally agreed upon measure of a recession.  We had explained that measure, and we had explained that the period we would live through would be “hopefully not a recession, but still really lousy.”  Technical measures aside, by most common sense definitions, we are in a recession that started in either December 2007 or January 2008. 

Still, we are not predicting a technical recession.  GDP growth for the fourth quarter of 2007 was a positive (not a negative) 0.6%, consistent with Berkshire Money Management’s forecast.  Available economic data for this quarter is so far tracking similarly flat GDP growth (+/- 1%).  Fine tuning the forecast would be a mistake, however.  Although the quarter is almost over, only about half the data that enter directly into the GDP estimate have been released.  The good news is that the January trade deficit was measurably better than expected.  The bad news is that, especially during downturns in the business cycle, downward revisions are common.  (As of right now, if I was forced to, I would argue that Q1 2008 GDP growth, at best, duplicated last quarter’s growth.)

There is little doubt that the U.S. economy has, at best, stagnated and very possibly could still contract.  Employment and vehicle sales are falling, retailers are reporting slumping sales, business and consumer confidence is flagging, and unemployment insurance claims are on the rise.  The negative, self-reinforcing dynamic that characterizes recession –higher unemployment undermining consumer confidence and spending, prompting firms to cut payrolls in turn, and adding further to unemployment – appears to be in full swing.

As indicated earlier, although the traditionally agreed upon measure of a recession may not be reached, by most common sense definitions we are in a recession that is now three or four months old.  Given the scenario that a recession is underway (for the sake of fluidity in this report, we will refer to the current economic scenario as a recession), that does raise the question of what we can thus expect from the stock market. 

It is important to note that the average duration of the ten recessions that have occurred since World War II has been 10-months.  True to form, the stock market is anticipatory of business cycle movements.  The stock market tends to peak about 7-months before the recession starts (that would have been, on average, May/June of 2007 if the recession started in December/January). 

The average duration of a stock market slide during a recession is about 12-months.  The stock market peaked at very similar price points in both July 2007 and October 2007, and thus an average duration of stock market pain would last until July or October 2008.

As Chart S01724 exhibits, the stock market, on average, anticipates an improving economy 5.5 months from the start of the recession.  That would suggest a stock market bottom in April/May 2008 (perfectly consistent with our Economic Outlook 2008 report suggesting stock market weakness through the first four or five months of the year). 

An important note to the past historical action, about 2.5 months into a recession (so February/March) you see a quick “double-bottom” of the variety we have recently experienced, followed by a quick rally, and then ultimately lower lows.  It seems very clear that the stock market will rally over the next 2-6 weeks – what is not yet clear is what the ultimate low will be for the stock market this year.  Historically speaking, any rally from here may be perfectly sustainable for the next year, but that does not mean that the lows for the year have already decisively been hit.

On the subject of double-bottoms, please allow me to define what I am referring to.  In the beginning of the year Berkshire Money Management defined the risk level of the S&P 500 to be 1,300 points.  We rapidly hit that level in 14 trading days.  We quickly posted an article on the website explaining that we hit a bottom and were going to get a very good rally.  And that rally materialized to the tune of 1,267 points on the Dow.  However, we then just as quickly posted another article on the website explaining that we did not like the internals of that rally and that we should expect a re-test of the recent lows, back to our pre-defined risk level.  We got that re-test and the chart formation depicts a “double-bottom.”  That is a bullish sign in technical analysis.

The thinking behind the double-bottom concept is that if buyers surface twice around the same area (in this case, 1,300 points), it creates a recognizable floor for the fair value of stocks.  It indicates a price level where investors have been consistently comfortable buying into stocks.

Also of interest, March 2008 marked the month in which it was announced there had been two back-to-back months of negative non-farm payroll growth.  In four of the last six recessions, the month in which that announcement was made marked the bottom of the stock market’s descent.  The two exceptions were 1982 and 2001.  In 1982, the market bottomed several months later, but at levels not far off from the previous levels.  In 2001, the market hit bottom much later as it worked off excessive valuations.

Should our 2008 Risk Level of 1,300 points for the S&P 500 continue to hold (as opposed to a much truer level of capitulation) then the available forecasting tools that have existed since stock markets themselves have existed will again prove to be too blunt to pinpoint any turning point with great accuracy combined with great probability of success. (In contrast, when Berkshire Money Management called the stock market bottom to the day in October 2002, the tools were much more precise due to the high levels of fear and capitulation.  I know that it does not feel like it, but this has so far been far too ordinary a business-cycle led stock market decline to have sharpened those available tools.)

Whatever, and whenever, the ultimate low point for the stock market in this cycle, it is important to note that recessions lead to incredible buying opportunities.  It is important to note that the demarcation from a bad market to a good market will not be clear nor will it be V-shaped.  The bottoming process will involve a transition period of probing and testing followed, ultimately, by a significant reversal of major trends. 

From the stock market’s low point of the last ten recessions, the S&P 500 has been up an average of 16% three months later, 24% six months later, and 32% a year later. 

—% Gain, Trading Days Later—

S&P Low Date
In Recession		63	126	189	252
06/13/49		14.5	19.2	26.6	33.7
09/14/53		9.9	17.7	27.5	38.5
10/22/57		6.1	9.8	19.0	31.5
10/25/60		15.9	25.2	27.6	30.9
05/26/70		16.9	20.8	38.7	44.5
10/03/74		13.5	29.9	51.5	34.6
03/27/80		18.3	31.1	39.1	37.1
08/12/82		37.8	41.6	61.1	61.1
10/11/90		6.7	28.8	28.7	28.8
09/21/01		18.0	17.2	2.8	-13.7
Mean		15.8	24.1	32.2	32.4
Median		15.2	23.0	28.1	34.2

What happened in years that included not only a bear market and a recession, but also a presidential election?  That was the case in 1960 and in 1980.  Both were years in which the incumbent party lost the election.  As the table illustrates, in both cases the market performed similar to the average gains of the other years over the measured time periods.

In our Economic Outlook 2008 report, we separated calendar year 2008 into the first third and the second two-thirds.  We argued that the first four or five months would be weak for the economy and weak for the stock market.  We explained that if our forecast for the first part of the year held true (and unfortunately it has) then that would bolster our conviction of our forecast for the second two-thirds of the year. 

Our forecast for the second two-thirds of the year was that the stock market would turn upward in anticipation of the economy gaining traction due to the massive amounts of fiscal and economic stimulus injected into the economy. The market’s recovery halfway into recessions has a lot to do with changing expectations.  Ahead of a recession and during its initial months, the market tends to doubt that Fed rate cuts can do anything to save the economy.  The market tends to be focused on dropping earnings growth and the apparent slim chances for earnings to recover anytime soon.

Given these split forecasts, the sectors to favor for the first part of the year would have been Health Care and Consumer Staples, as these defensive areas have fared well during weak markets.  For the second part of the year Information Technology and Consumer Discretionary would be in favor.  That forecast continues to stand.  You can add Financials and Small Capitalization stocks in general to that short list of more favorable areas for the second part of 2008.  As you can see in Chart AA0148, in the same way that the stock market tends to recover six-months into a recession, that is also when small-caps tend to start outperforming large caps.

Should we actually be in a technical recession (as opposed to just a “common sense” recession), it could very well be the shallowest recession on record.  Let us consider that a recession started in December 2007, we now have a full three-months worth of data to which we can compare to the average of the first three-months of previous recessions.

	Avg. 1st 3 months of last 2 recessions	Latest Reading
Weekly Jobless Claims	387,000	374,500*
Job Losses	114,000	80,000
ISM Index Manufacturing	44.1	48.6
ISM Index Non-Manufacturing	49.3**	49.6

*4-week average (jobs not adjusted for population inflation)

** 2001 recession only

The Institute of Supply Management’s (ISM) national manufacturing index has fallen to 48.6, a level indicating that the factory sector is slightly shrinking (below 50 is shrinking, above 50 is growing).  But the ISM says the index needs to fall to about 41 to indicate a broader recession.  The 4-week average of Weekly Jobless Claims has been creeping higher, but it is still less than the roughly 400,000 level (adjusted for population inflation) breached in prior recessions.

Consumer spending, 70% of the economy, was nearly flat, when adjusted for inflation, in the first two months of the year.  That’s not good, but it might not be bad enough to turn first-quarter GDP growth negative, especially with a weak dollar propping up exports.

The ten previous recessions since World War II have varied widely in length and severity.  On average, real GDP (nominal GDP minus inflation) declined peak-to-trough by nearly 2% and employment fell by 2.5%.  The last two recessions, in 2001 and 1990-1991, were particularly short and mild, each lasting only eight months, with GDP falling by less than 0.5% and unemployment rising by 2.5 percentage points. This recession is expected to be short and mild; no worse than the 2001 downturn.  The 2008 recession is thus expected to be the mildest downturn of the post-World War II period. 

The heated primary season has brought election-year tendencies to the forefront.  The S&P 500’s -6.1% decline in January was the fifth worst since 1950.  The S&P 500 has declined six times during Januarys of election years (1956, 1960, 1968, 1984, 1992, and 2000).  Chart 0804A_C, shows that the average of those six cases has been weaker for the stock market than the average of all election years, and that the market has tended to be weak early in the year before starting a choppy uptrend.  This might be expected given that a study of the so-called January Barometer which shows that when January has been up, the S&P 500 has risen 12.3% on average from February through December.  Conversely, when January has been down, the S&P 500 has declined -0.9% on average over the next 11 months.

In all but two of those six times (1956 and 1984) in which Januarys declined during election years, the incumbent party lost the presidential election.  In our Economic Outlook for 2008 report, and as Chart Davis520 illustrates, we pointed out that the market’s election-year performance has been worse when the incumbent party has lost, and that the majority of the weakness has occurred during the first half of the year. 

Election years are notorious for volatility, with many min-rallies and mini-corrections. As the following table shows, the average of the largest of the corrections during an election year (the “greatest correction”) has been -13.3%.  And the largest of the rallies during an election year (the “greatest rally”) has been an average +23.4%. The historical context suggests that there is more reward potential than downside risks from current levels.

For example, from the beginning of the year we might have expected the S&P 500 to fall from 1,468 points to 1,272 points by the third week of March (sound familiar)?  The resulting rally would bring the S&P 500 to 1,570 points by in August (but temper your expectations, as there are valuation and recessionary headwinds that you would not find in typical election years).

Election Year Rallies and Corrections

	Greatest Rally %	Average # of Rally Days	Greatest Correction %	Average # of Correction Days
Mean	23.4	152.2	-13.3	79.3
Median	16.5	152.2	-9.9	53.0
Incumbent Party Wins
Mean	21.5	169.6	-10.7	77.6
Median	16.5	207.5	-9.9	48.5
Incumbent Party Loses
Mean	26.2	126.1	-17.1	82.0
Median	16.8	139.0	-12.6	68.0

• Follow up to the January 23rd website article “When Everyone is Looking One Way”

In that article we explained the rationale behind why we thought that the stock market would rally from that point, which it did over the next week to the tune of 1,267 points on the Dow (ultimately we did not like the internals of that rally and said so in our January 31st article warning of a re-test of the January lows, which obviously happened).   The idea behind the article is that when everyone is looking one way (when everyone is very bearish), you might want to consider going the other way (getting bullish). 

There are a number of indicators that measure optimism and pessimism.  To spare you the details, let me just say that to find numbers as comparably bearish as some of the current available indicators you have to look back to October 2002.  The stock market turned up quickly from that point, surging 10% in just four weeks.  The stock market later fell back toward those lows by March 2003, but over the long run it turned out to be an excellent buying opportunity.

• Follow up to the February 11th website article “Revisiting the Risk Level”

In that article I cited several reasons explaining why Berkshire Money Management believes that we are working through a bottoming process. One of those reasons was bullish insider buying activity.  The 2008 insider data remains positive, suggesting that corporate insiders remain optimistic for more market gains (if not, they would be moving to the sidelines). 

Late December saw the start of a sharp contraction from heavy selling they were doing in previous months, a bullish sign.  The market’s January slide saw insiders increase their buying and further halt their sales.  For two weeks corporate insiders actually bought more shares than they sold.  That rarely occurs except after a large decline, and that was the situation on January 22 & 23.  The more recent data reveals fewer than 1.2 sales for each purchase.  It is bullish for the ratio to be below 2.0, and it has been in that bullish zone since December 24th. 

2009 & 2010 RISKS

• Despite which party wins the White House, there is virtually no possibility of new legislation that will halt the expiration of the Economic Growth and Tax Relief Reconciliation Act of 2001 (aka The Bush Tax Cuts).
• Due to the expiration of the Bush Tax Cuts, the stock market will face demanding headwinds as the investor class becomes troubled with higher tax rates on dividends and capital gains.
• Despite which party wins the White House, tough decisions will have to be made in 2009 and 2010.  These tough decisions will be a drag on the economy as the new administration will “clean house” and will defer economically-friendly decisions until the next election cycle approaches.

Political commentary aside, the December 31, 2010 repeal of the Economic Growth and Tax Relief Reconciliation Act of 2001 (aka The Bush Tax Cuts) is going to a big negative to the stock market.  Starting in 2011 the tax rates for both dividend payments and capital gains are going to jump.  It seems painfully obvious (painful to investors, at least) that no matter which party wins the general elections, these cuts will be repealed. 

Bipartisan government think-tanks (I know, an oxymoron, right?) have measured that the positive effect on stock prices was, at the time, worth about 800 points on the Dow, or a ten percent gain.  There is no historical context to make comparison, but it seems fairly obvious the if the introduction of the tax cuts were good for a ten percent gain, then the repeal of the cuts would be good for a ten percent loss.  That, in itself, is not a big deal.  Historically speaking a ten percent drop happens almost once per year – so no big deal. 

But within the context of the repeal coming as we muddle through the first two years of a new administration making difficult choices (as opposed to the more stimulative decisions typically made in the final two years as government officials prepare for re-election), that ten percent drop could easily be magnified. 

But when?  That is the problem.  There is no historical context to which we can suggest that the stock market will begin pricing in a tax increase on investors.  Does the stock market discounting begin to occur in August during the Democratic convention (as we may find out then just who the Democratic nominee may be)?  Or does it happen in November after the general election?  Or does it happen in 2009 as the President starts to make difficult, non-stimulative decisions?  Or does the discounting hold off until 2010 when investors realize that it may be to their benefit to start reallocating their stock positions?

Given the lack of tools available to decipher the timing of stock market anticipation of new and negative legislation, we are likely to keep the use of Absolute Return CDs in our arsenal for Schwab accounts, as not only are they FDIC-insured, but they allow for market gains in both up and down markets.