On Friday evening, the NASDAQ share index was down 59.3 percent from its peak, set on March 12, 2000. But in spite of all the economic gloom, the Dow Jones index over the same period was up 7.2 percent.

Was the bubble confined entirely to the New Economy, or is there another shoe to drop?

To investigate this one, the Bear is going to have to do some actual work. The only way, apart from guesswork, to determine what should be the appropriate level of the Dow Jones average is to look at the shares contained in it, and see where they should be selling.

As I’m sure readers know, the Dow Jones Industrial Average, which was established in 1896, is an un-weighted average of the share prices of 30 stocks. To get the average, the share prices are added up on a simplistic basis, and then divided by a divisor, which changes each time new stocks are included in the average or companies undertake a stock split. This divisor, having presumably in 1896 been 12 (the number of stocks in the then average) is now 0.15369402, the decline reflecting how many stock splits there have been in 105 years of a generally rising market.

Rather than analyze all 30 stocks, which would produce a very long and tedious column, I will confine the analysis to those 10 stocks with the highest prices (and consequently the largest weights in the Dow) which together account for almost half of the index, and assume that the other half’s valuation adjustment is more or less similar.

In normal times this might introduce a bias since those stocks which had sunk would be of lower price. But today, after the huge bull market, this is not the case because 25 of the 30 Dow stocks have split in the last five years. So which stocks are priced higher reflects mainly the timing of splits, and is thus random.

The ten stocks concerned, in declining price order as of Friday evening, are Minnesota Mining and Manufacturing (3M), IBM, Johnson & Johnson, Exxon Mobil, United Technologies, Merck, Procter & Gamble, Boeing, General Motors and Microsoft, a pretty fair cross section of U.S. industry, with a couple of large New Economy names reflecting technology’s continuing share in U.S. business.

To analyze the companies, I will postulate a medium sized recession, with a GDP drop of something between the 1.3 percent of 1990-91 and the 3.4 percent of 1974-75. I will also assume what I have argued to be the case, that the “productivity miracle” of 1995-2000 is largely a statistical blip, that total factor productivity growth remains close to its long-term average, and that required returns on stocks are not those of the 1973-82 bear market, but are close to those prevailing in 1991, a relatively positive period for the market in spite of the recession (on March 12, 1991, the Dow closed at 2922.52, 6 percent above the bull market peak of 1987, and 60 plus percent above the bottom reached after the crash of that year.)

From this data, I will project a stock price for each company in March 2004, three years from now, based on 2003 earnings, assumed to be at or just after the nadir of the 2001-(whenever) recession.

A particular problem relates to stock options, which provide remuneration to company staff without appearing in the income statement explicitly. These were a minor feature of the scene 10 years ago, so earnings comparisons between 1990-92 and 1999-2000 are distorted if they are excluded.

As discussed in previous columns, the cost of stock options exercised can be calculated independently of the Black-Scholes model, and is the number of options exercised multiplied by the profit to the holder between the option strike price and the stock price at time of exercise.

I will thus look at the stock option cost per share on this basis in last year’s earnings, and will deduct one quarter of it, a conservative proportion reflecting the fact that 1999 and 2000 were good years for the market, with correspondingly high option gains for holders.

In some cases (notably Microsoft) this results in a markedly lower EPS figure because these are the companies which will have to pay their employees more in cash when the market drops, issue ever greater numbers of new options and diluting stockholder wealth, or suffer damaging staff defections, even in a recession.

Company by company, the analysis runs as follows:

— 3M, at $116.41 on Friday evening, is a very solid company which has achieved exceptional returns over the years through incremental innovation in a broad range of non-technology product areas, most famously including Scotch tape and Post-It notes.

Its operating sales margins appear to be more or less recession-proof, being 15.7 percent in 1991 and 17.8 percent in 2000. Its price-earnings ratio in 1990-91 was around 17 times. Losses to the company from stock option exercise in 2000 totaled $126.5 million, or $0.27 per share, one quarter of which is $0.07 per share. Thus 2004 earnings, assuming 3M’s usual steady 5.3 percent annual sales growth, and recession margins of 16.5 percent can be estimated at $1.931 billion, or $4.09 per share, less $0.07 is $4.02, multiplied by 17 is $68.34, down 41.3 percent from today’s price.

— IBM, at $99.29 Friday evening, is a highly cyclical company which has redefined its business from its historical position as master of the mainframe to a more generally strong position in the corporate computer hardware and software areas. Comparisons are difficult since the company lost money in 1991-93. Nevertheless, if we assume that sales three years hence would be about flat given a technology downturn that is already happening, sales margins in a recession would halve to around 4.6 percent from 2000’s 9.2 percent. The company’s price-earnings ratio in a recession would be 20 times, down from today’s 25 times but still reflecting some of the glamour of IBM’s sector rather than its anemic growth rate, then IBM’s recession earnings per share would be $2.22 fully diluted. 1999 stock option gains, the most recent available, were $2.015 billion, or $1.11 per share, so $0.28 must be deducted from the EPS, to give an EPS of $1.94 and a stock price of $38.80, down 60.9 percent from Friday’s level.

— Johnson & Johnson, at $96.70 Friday evening, is another steady growth company like 3M, specializing in health care and cosmetic products. Sales growth in 1990-2000 averaged 10 percent per annum, while pretax sales margins increased from 13.2 percent in 1991 to 15.2 percent in 1999. Stock option exercise losses were $684 million in 1999 or $0.48 per share, so $0.12 per share must be deducted from 2003 EPS. Assuming 8 percent sales growth into the recession and a 14 percent margin, 2003 EPS would be $2.62, or $2.50 after stock options.

At a price earnings ratio of 23 times, that gives a stock price of $57.50, down 40.6 percent.

— Exxon Mobil, at $86 Friday, is a merged integrated oil company which had a good 2000. Assume here that oil prices in 2003 are roughly the 1990’s average (since the recession will reduce pressure on supplies), then since the company is at essentially nil growth on an operating basis, 2003 earnings can be expected to be about $2.50 per share. 1999 option exercise cost was $536 million, or $0.16 per share, a quarter of which is $0.04 per share. At nil growth, Exxon Mobil is worth only 12 times earnings, so projected 2004 stock price is $2.46 x 12 or $29.52, down 65.7 percent.

— United Technologies, at $81.70 Friday, is in a number of cyclical businesses, including aircraft engines, elevators and air conditioning. Sales are increasing at around the nominal rate of GDP growth, primarily by an active program of acquisition and disposal, while operating margins in a recession are around half those in 2000. Hence expected 2003 EPS, assuming 6 percent sales growth, are $2.11 per share. Stock option profits in 2000 were $372 million, or $0.73 per share, so subtracting one quarter, $0.18 per share, gives $1.93 per share 2003 earnings. The multiple applied to these should be no more than 15 times, reflecting slow growth and high risk, giving a stock price in 2004 of $28.95, down 64.6 percent.

— Merck, $75.70 on Friday, like Johnson and Johnson is a very high quality company in the drug business (but more prescription drugs rather than OTC drugs.) For 1990-2000, sales grew by 18 percent per annum, primarily by acquisition, while pretax margins, 36.8 percent in 1991, had declined to 24.3 percent by 2000 and are clearly under substantial further pressure. Stock option exercise costs were $1.15 billion, or 48 cents per share, so $0.12 should be subtracted from 2003 earnings for comparability. On this basis, assuming 10 percent sales growth and 2003 margins of 20 percent, 2003 adjusted EPS would be $3.05 and the stock price, based on 25 times earnings for such a premier growth company, would be $76.25, up 0.7 percent from 2000. Clearly the standout current and likely future performer of the ten.

— Proctor and Gamble, $69.14 on Friday, is another steady company, based in household products, which after growing by about 8 percent per annum in the 1990s seems to have gone ex-growth in 1998-2000. Stock option costs in FY 2000 were an exorbitant $1.78 billion for a company with no earnings growth, or $1.36 per share, more than 50 percent of earnings, so deduct $0.34 from 2003 earnings for comparability. Assuming that a company that cannot grow earnings in 1998-2000 will find it hard to grow them in 2001-03, project 2003 EPS, adjusted, at $2.13, for a stock price based on 15 times earnings for a slow growth company of $31.95, down 53.8 percent.

— Boeing, $65.50 on Friday, is a highly cyclical company whose premier position in aircraft has been badly eroded by Europe’s government-subsidized Airbus in recent years, and which is selling at 23 times cyclical peak earnings. Sales growth can be assumed to be zero over the next three years, as the aircraft cycle turns down. Net Income margin was $2.88 per share in 2000, a cyclical peak, but a loss in 1995. Stock option exercise costs were $50.1 million, or $0.05 per share, i.e. almost negligible (the first of these!). Hence, assuming net income of $1.50 per share in 2003, and a price-earnings ratio of 20 times, down only modestly on today’s level, we get a 2003 projected stock price of $30, down 54.2 percent from today’s level.

— General Motors, $58.95 on Friday, is another highly cyclical company, in automotive and aerospace, that is already running into trouble, although 2000’s figures were good. The company is effectively ex-growth, and loses money in every recession. In 1999, GMs stock option exercise costs totaled $314 million, or $0.51 per share. Assuming 2003 margins are half those of peak year 1999, EPS would be $4.68. Subtracting $0.13 for options, and taking a 10 times price-earnings ratio for this cyclical, no-growth stock gives a 2004 stock price of $45.50, down 22.8 percent on Friday.

— Microsoft, finally, $56.69 on Friday, is the world’s premier software company, but with serious competitive and legal threats. Its margins have always been extremely high, and Net Income margin was 41 percent in the year to June 2000.

Sales and earnings growth are however slowing, revenue growth was only 8 percent in the quarter to December 2000 over the corresponding quarter of the previous year, and net income in the year to June 2001 is expected to be up only marginally.

Stock options exercise cost in the year to June 2000 was a colossal $13,925 million, or $2.64 per share, so $0.66 per share must be subtracted from 2003 earnings for this dilution. Assuming a continuation of 8 percent per annum revenue growth and a 30 percent Net Income margin, earnings in the year to June 2003 will be $1.58 before stock options dilution, or $0.92 after dilution. Microsoft’s price-earnings ratio will undoubtedly take a hit from such lackluster performance; even so, hope springs eternal, and we can project a ratio of 25 times earnings in 2004, for a stock price of $23, down 59.4 percent.

Adding up the ten stock prices above gives a total of $452.31, grossing up for the other 50.74 percent of index components, and dividing by the divisor of 0.15369402 gives a March 2004 Dow Jones Industrial Index of 5,676.

Phew, first.

The figure of 5,676 obtained hereby is actually somewhat above my own expectation of the bear market bottom, which would be in the 3,500-4,000 range. However, the assumptions herein are relatively conservative. In particular, the assumption of only a moderate recession does not take account of the loss in stock value, measured by the Wilshire Index, of about $9 trillion, if this calculation is correct, i.e. 90 percent of current GDP.

While lower than Japan’s 1990-2000 wealth loss of about 200 percent of GDP (which figure includes golf course memberships and urban land, as well as stocks) this still suggests that the oncoming recession, driven as it will be by asset rather than inventory effects, will be quite spectacularly painful.

-0-

(The Bear’s Lair is a weekly column that is intended to appear each Monday, an appropriately gloomy day of the week. Its rationale is that the proportion of “sell” recommendations put out by Wall Street houses remains far below that of “buy” recommendations. Accordingly, investors have an excess of positive information and very little negative information. The column thus takes the ursine view of life and the market, in the hope that it may be usefully different from what investors see elsewhere.)

This article originally appeared on United Press International.