Insanity in the Markets

Whether or not the ostrich (the investing public) has its head in the sand, the insanity above ground level still exists. Should we be taking responsibility for the position of the ostrich's head? Probably not. I will therefore understand if you decide not to publish this. I wrote it as a reality check on my own sanity. Where I got to in my thought processes is that I am a sane individual sitting at the mad hatter's tea party table.

Microsoft Mirage

Over half a lifetime, my interest in equity markets has been oriented mainly to human behaviour, and the extent to which the human mind is capable of rationalising that which - in hindsight - begs the question: "What in God's name were we thinking?"

This orientation was initially stimulated by my wonderment at the behaviour of otherwise reasonable German citizens prior to WWII who, years (and even generations) following WWII, were still themselves wondering what had come over them. Why do "normal" people behave in a manner that defies all rationale? What makes what is objectively insane a socially acceptable frame of reference?

The best way to explain the relevance of this question in November 2004 is by pointing to an example.

There is no doubt that Microsoft has been the wonder stock of the last 20 years or so. It's closing price on Friday November 4 th 2004 was $29.00 a share, and this represented a P/E ratio of 36.5X on trailing earnings according to yahoo.com; and a P/E of 22.9X on forward (projected) earnings.

Now, let's get some perspective on this. On Friday, November 4th, 87,894,440 shares changed hands - which represented a dollar turnover of around $2.5 BILLION in a single day. Further, at $29.00, the share is trading at close to the high of its 52 week range. I will demonstrate below that this represents 25% of the company's real net worth that was traded in a single day. Supposedly sane and rational investors were prepared to put $2.5 billion on the table in a single day to invest in the Microsoft business.

Let's look at some fundamentals of Microsoft:

First, its Net Tangible Assets as at June 30th 2004 were $71.1 billion. There were 10.9 billion shares in issue - which implies that each $29.00 share was backed by $6.52 of net tangible assets.

Importantly, of the $71.1 billion NTA, $60.5 billion was in the form of cash and short term investments.

Now, just to emphasise the point, let's assume that the $60.5 billion is fully used on November 6th 2004 to repurchase shares at $29 each. This means that the number of shares in issue would drop by 208 million, leaving 10.69 billion shares in issue, with net tangible asset backing of $10.6 billion. i.e. Each $29 share would have tangible asset backing of just $1 (that's on dollar!).

OK, so it's clear that people are not investing in Microsoft to acquire assets.

So let's look at its earnings power. Microsoft is clearly a "cash cow" that has huge operating margins. In fact its ratio of "Operating Profit:Revenues" was roughly 25% in 2004.

Now, when we look at the "Operating Profit" line we see that it was $9 billion for the year to June 2004 (down from $13.2 billion the year before).

And the question arises: How much of this "Operating Profit" was Interest Income earned on the $60.5 billion cash pile?

Most companies highlight interest paid AND interest earned - to enable analysts to calculate "Earnings BEFORE interest and Tax" (EBIT), but it is noteworthy that Microsoft's Profit and Loss Statement shows ZERO interest paid AND it does not make any reference at all to Interest EARNED.

Logic therefore dictates that the interest earned component of income is buried in Operating Profit. The reason this analyst arrives at this conclusion is that "other" income was shown as $3.16 billion, and after this has been ADDED to Operating Profit, the total is described as EBIT. i.e. Microsoft declared in the year to June 30th 2004, a total Earnings BEFORE interest and tax of $12.2 billion.

Now, let's play a game here. Let's assume that the items in the balance sheet entitled "Cash and Cash Equivalents" and "Short Term Investments" were invested in 30 year Government Bonds at the prevailing yields. As a dividend yield of 5% is not strictly "interest", the accounting standards would have been technically complied with. Of course, the fact that the bonds may or may not be described as 30 year instruments is irrelevant to their categorization in the Balance Sheet. Because they can be traded (sold) at will, they could be classified as Short Term Investments.

5% (the average prevailing yield on 30 year bonds in 2003/4) of $61 billion = $3 billion of the $12 billion "EBIT" in 2004 may have been accounted for by dividends received (and very possibly it was actually categorised as "other" income).

But in 2003, the bond yields were also travelling at around 5%, as can be seen from the chart below:

This implies that if the full $49 billion cash etc shown in the 2003 Balance Sheet was invested at 5%, then "other income" would have been around $2.5 billion in the 2003 year.

In fact, other income in 2003 was only shown at $1.5 billion, so its more probable that the dividends were categorised as Operating Profit. Whatever the actual categorization was is really irrelevant. What IS relevant is that we need to make some adjustments to 2004 income to see what earnings would have been if the cash in the bank had in fact been used (according to the rules of our imaginary game) to buy back the shares a year earlier.

Because the Balance Sheet would have had $60 billion less assets, there would have been no dividend income. Income before tax would have fallen to around $9 billion and, at 33% tax rate, the after tax profit would have been $6 billion.

The eps on the now lower number of shares would have been $6 billion divided by 10.69 billion shares - or 56 cents a share, and the trailing P/E ratio (excluding financial "humbug" ) would have been $29/$0.56 = 51 X

Of course, we cannot ignore the fact that existing shareholders would have been enriched by the proceeds of the share buy back. They would certainly have their cash in hand (on which they could also earn 5% just by parking it in Government Bonds) and they would be left with shares with a trailing P/E ratio of 51 X.

"Aha!" I here the professional analysts bellowing. "But you forget that forecast earnings are much higher than trailing earnings".

OK, let's assume that EBIT gets back to what it was in 2003, viz $14.7 billion. From this we can deduct our imaginary yield on cash and short term investments of $3 billion, and the net result will be $11.7 billion before tax or $7.8 billion after tax. On adjusted outstanding shares of 10.69 billion this would represent $0.72 and the FORWARD P/E would be $29/$0.72 = 39.7 X - i.e. It would take roughly 40 years to get your capital back unless Microsoft's earnings continued to explode.

Are we missing something here?

Well, an interesting item on Microsoft's P&L is expenditure on Research and Development - which jumped from $4.6 billion in 2003 to $7.7 billion in 2004. Clearly Microsoft is pulling out all stops to design a new suite of products.

What could these products be? To be truthful, neither I nor (probably) anyone else could say, but let's keep playing the game. On a "best case" scenario, Microsoft will somehow miraculously develop a new suite of products which are just as successful as their existing suite and so the profitability will DOUBLE from $7.7 billion to $15.4 billion - assuming their revenue doubles to $73.6 billion. This is an "insane" assumption because no product will go from zero to full market penetration in one year. Of course, there will be some naïve analysts who will argue that all incremental revenue will generate profits that will fall straight to the bottom line on a gross to net basis. Let me assure these people that this will not happen. The ratio of EBIT:Sales will not rise. What will happen is a whole swag of new overheads will need to be incurred to provide infrastructural support to the new sales. Expenses such as advertising, help desk, administration, software maintenance engineers, sales commissions, etc etc. will have to be added.

Even if our "wild assumption" of $15.4 billion materialises, this will represent eps of $1.44 and a PROJECTED P/E - on a business that it is assumed will have reached 100% of its potential in one year - of $29/1.44 = 20 X

It will still take 20 years for investors to get their capital back - assuming that a business which is already the dominant player in its existing mature markets becomes an overnight dominant player in a "new" and instantaneously mature market.

Are you starting to get the hang of this yet?

What we are dealing with in November 2004 - purely and simply - is "insanity" on an Investment Community Wide scale. The probability that Microsoft will double its profits overnight is "vanishingly small", and the fact is that whatever their new product/s is/are, there will be risks associated with whether or not the market will buy these new products with enthusiasm. There will also be questions regarding the potential size of these new markets. What is the probability that the new markets will even be the same size of the existing markets? What is the probability that Microsoft will dominate these new markets?

So we are talking about a "best case" scenario leading to an "instantaneous full market penetration" and an "instantaneous" realisation of full market share and full "profit potential" leading to a forward P/E ratio of 20X if you back out the huge pile of cash that is sloshing around on the Balance Sheet - and which appears to be dazzling the investment analysts to the point that their brains have become addled.

What are the attaching probabilities? In simple English: Vanishingly Small.

So where does this leave us?

It leaves us smack bang in the middle of an "Investment Mania" that has been ever so slowly deflating since 2004 but is still very much present and accounted for.

Now let's look at Microsoft's long term chart (courtesy Bigcharts.com) to see whether it is offering any clues regarding where this insanity might end.

One possibility is a level of $18-$20, if one draws a trendline by joining the low of 1992 to the low of 1995, and extending it to 2005.

Another possibility is that the share could come all the way back to between $5 and $12.

A further possibility is that the shares could trade sideways to mildly up for upwards of a decade before the P/E ratio comes back from the land of Nod as earnings grow over a protracted period via a combination of market penetration and price inflation.

Yet another possibility is that the Mad Hatters Tea Party can continue for some time to come and, like Google, Microsoft's shares can continue to rise like a helium balloon. We are, after all, not dealing with a "rational" world here.

Conclusion

In the land of the blind, one eye is king. This "back of the envelope" analysis - which did nothing other than assume all surplus cash is backed out, and that earnings double overnight - provides a clear picture of investor insanity which, in years to come, will beg the question: "What in God's name were we thinking?"

But the question is one of timing. When will reality dawn?

The following chart of the Dow Jones is showing that price has now reached the resistance of its downward sloping trendline (courtesy DecisionPoint.com):

The fact is that most "short term" trading indicators are now horrendously overbought. Eg, well over 90% of individual PMO's are rising, and this state of affairs typically does not last more than a few days.

But if the Dow Jones pulls back from its Friday level, it will be pulling back from yet another high that was lower than its previous high, and it will remain below its falling trendline.

And if it does this, then it is highly likely that the MACD on the monthly chart is likely to give a "sell" signal within the next thirty days.

Overall Conclusion

The "Indian Summer" that I referred to all those months ago, now appears to be drawing to a close.