Leithner Letter No. 19
26 July 2001

In general, the practices advocated [by the Securities Industry Association in the U.S.] are fine as far as they go, useful for encoding certain minimum procedural standards that could prevent the very worst abuses. But the industry’s effort leaves plenty to be desired. Most important, the guidelines don’t address the tougher, more substantive problem with research as it’s generally performed today: Too often analysts lack broad and deep understanding of industries and fail to undertake the rigorous independent legwork to divine the true prospects of the companies they follow. Just as crucial, many don’t employ real discipline in valuing companies, the very thing investors most sorely crave.

Michael Santoli
“Proposed Guidelines for Wall Street Analysts
Won’t Solve Their Credibility Problem”
Barron’s 18 June 2001

Surprise Surprise

Telstra Corp. Ltd is one of the largest companies listed on the Australian Stock Exchange. It is also one of the few that is big enough to attract the attention of American, British and other major international institutional investors. The Commonwealth Parliament owns slightly more than 50% of its shares; further, Telstra once held a statutory monopoly over – and today services approximately two-thirds of – the Australian telecommunications market. Hence it is subject at least as much to party-political and regulatory forces as it is to commercial pressures. It also has one of the country’s widest and deepest share registries: accordingly, most Australian portfolios (Leithner & Co.’s is an exception) hold Telstra shares. 

For these reasons, Telstra’s announcement to the ASX at 09.07 on 12 June, like its other market-sensitive announcements, attracted the attention of scores of politicians and journalists, hundreds of institutional market participants (advisors, analysts, funds managers, etc.) and hundreds of thousands of individual shareholders. 

The first paragraph of the announcement stated: “with trading results to hand for the month of May, and year to date, Telstra. reiterates public commentary by management relating to slowing industry growth and now confirms that [its] revenue and profit growth for the second half of the 2000/01 year will fall below historic trends.” The announcement concluded that “management’s projections foreshadow that the year ahead will see growth in operational free cash flow, a continuing strong balance sheet and credit rating, and the expectation of increasing margins on the back of a new generation of cost reduction efforts.” At the close of trade on 12 June the market price of Telstra’s shares decreased by more than 10% and its market capitalisation fell by almost $A4 billion to approximately $A38 billion. Further falls, depressing these measures to three-year lows, occurred in the second half of June.  Telstra’s announcement surprised most journalists and analysts. Robert Guy, for example, commenced his report (Australian Financial Review Weekend Edition 16-17 June) with the words “Telstra saw $5.7 billion wiped off its value during the week after delivering a surprise profit warning on Tuesday.” He added that “the surprise profit warning saw leading analysts cut back their investment recommendations on the stock” and that “analysts, who have a strong aversion to nasty surprises from leading blue chips, could barely hide their disappointment.”

Reading Between the Lines

Guy’s lead sentence is thoroughly orthodox. Very few academics, advisors, analysts, brokers, funds managers or journalists would query its premise that the market price and the value of a financial security are synonyms; equally few, therefore, would give a second thought to the notion that a change of market capitalisation implies a corresponding change of value. The conviction that at any given point in time price and value correspond comprises a cornerstone of the Efficient Markets Hypothesis (EMH) to which virtually all Australian market participants – amateur as well as professional – pledge conscious or implicit allegiance. 

In sharp contrast, a Graham-style value investor regards price and value as distinct phenomena: price is paid and value is received, and one does not exchange cash for a security unless one has grounds to believe that one thereby receives more in value than the price one pays. The value investor also expects that over time market cap and value gravitate towards one another. Accordingly, at any given point in time and possibly for extended periods of time price and value may diverge (sometimes by a wide margin). To the value investor, the underlying value of a business may therefore exceed its market price; conversely, price may exceed value. Hence the market capitalisation of Telstra fell by billions of $A during June; whether its value did so is an entirely different question – and one which movements of its share price cannot answer. 

To the mainstream, then, price equals value. By this definition the one cannot gravitate towards the other, and for this reason no ‘regression to the mean’ occurs on financial markets. Conversely, to the value investor price and value often diverge in the short or even medium terms – but in the long run there is a regression (of price) towards the mean (of value). 

A number of studies indicate that regression to the mean or something akin to it occurs on financial markets. Using data extending back over several decades, Richard Thaler and Werner De Bondt conclude that “extreme returns of stocks listed on the New York Stock Exchange were. subsequently followed by significant price movement in the opposite direction.” In consequence, if at a given point in time investors are unduly optimistic about a particular company’s securities, then (even when that company’s results remain basically unchanged) at some point in the future that sentiment is likely to reverse – and impact precipitously and adversely upon its market capitalisation. The simple but profound notion of regression to the mean, formalised by Sir Francis Galton in the nineteenth century, was recognised in rough outline at least 2,000 years ago. Today, however, it appears either that market participants have forgotten this principle or that their EMH-inspired methods obscure it.

‘Analysts’ Caught Napping

Table 1, which shows Telstra’s actual and estimated earnings per share (E.P.S.) for the ten-year period 1996-2006, illustrates this point. The second column shows its actual E.P.S. – including a ‘consensus estimate’ for 2000-2001. Assuming that Telstra earns $0.31 per share during the year ended 30 June 2001, its earnings will have compounded at a rate of 13% per annum since 1996. 

Table 1: Telstra’s Earnings Per Share
Actual, Exuberant and Chastened

Year Actual
Assumptions of 1999 (17%)
Assumptions of 2001 (7.5%) 
1996 $0.17    
1997 $0.27    
1998 $0.23    
1999 $0.27 $0.27  
2000 $0.31 $0.32  
2001 $0.31 (est.) $0.37 $0.31
2002   $0.43 $0.33
2003   $0.51 $0.36
2004   $0.59 $0.39
2005   $0.41  
2006   $0.44  

The table’s third column sets out the implications of an aggressive assumption which raised few eyebrows during the heady atmosphere prevailing in 1999: for the period 1999-2004, Telstra’s E.P.S. would grow at a compound rate of 17% per annum. (Such an assumption, or an even more aggressive one, was necessary in order to justify the purchase of Telstra at prices prevailing in 1999. The gulf between the reality in column 2 and the ebullient assumption in column 3 is immediately apparent: the assumption implied that Telstra’s earnings for the 2000-2001 FY would be 20% greater {[(.37 – .31) ÷ .31] = 0.20} than today’s consensus estimate. 

In sharp contrast, the table’s fourth column makes explicit the severely-chastened assumption implied recently by Robert Gottliebsen (The Weekend Australian 16-17 June 2001). Gottliebsen states that Telstra’s executives must increase EBIT (earnings before interest and tax) at a compound rate of 10% during the next several years, and opines that there is a strong likelihood that they will do so. (Indeed, reflecting analysts’ and journalists’ perpetually-sunny disposition, Gottliebsen states that “the long outlook for Telstra is now looking better than it has been for some time ... [that] the underlying value in the stock has never been better and institutions and individuals that have a long-term view are going to accumulate the stock at around these [$A5.65 per share] levels.”) Assuming that Telstra will pay tax during this interval (if it doesn’t it will disappoint the Australian Taxation Office), and that it will pay interest on its ca. $A7.0 billion of long-term debt (if it doesn’t it will default on this debt and thereby disappoint its bondholders), Gottliebsen implies that Telstra’s earnings after interest and tax (and, to be even more generous, ignoring hefty depreciation and amortisation charges) will increase at an annual compound rate of ca. 7.5%. 

The contrast between the exuberant and chastened assumptions is stark: by 2004 the exuberant assumption which raised no eyebrows in 1999 implies that Telstra’s earnings will be 50% greater than those implied by today’s chastened analysts and journalists. Similarly, the exuberant assumption led one to expect that Telstra would earn approximately $0.44 in 2002; under the chastened assumption, however, this level of earnings is not expected until 2006.

Note that nothing in Table 1 implies that either Telstra or the Australian telecoms market it dominates have changed or will change during these years. Rather, a stubborn and unco-operative reality (column 2) has obliged analysts’ bright expectations about Telstra’s earnings (column 3) to yield to comparatively dour expectations (column 4). Perhaps Telstra’s announcement of 12 June was “surprising” in the sense that it triggered analysts’ belated realisation of their error and its magnitude. If so, then given that very fashionable stocks and market segments, like Icarus, eventually return to earth, the intermittent fall of the price of Telstra’s common stock since 1999 (when it averaged $8.25 and on two occasions exceeded $9.00) can be explained largely in terms of a regression to the mean, i.e., the gradual reduction of its market capitalisation to a level which more closely approximates a plausible and justifiable appraisal of the underlying value of the enterprise.

Only tangentially, in other words, has the intermittent fall of Telstra’s market capitalisation since 1999 anything to do with the telecoms industry, Telstra’s management and operations, its combined state-private ownership, the vigour of economic conditions in Australia, etc. This interpretation, if it holds water, is greatly embarrassing to advisors, analysts, brokers and journalists. It might thereby explain their strenuous efforts during the latter half of June to devise ‘blue chip cover stories’ to distract attention from their error.

Sobering Implication No. 1 Telstra’s announcement need have surprised no market participant who was alert to the possibility that prices gravitate towards values; possessed widely available information; used this information to estimate value in a cautious, rigorous yet logically-elementary way; and interpreted the results of his analysis in a manner which avoided both Mr Market’s mood swings and institutional and bureaucratic imperatives.

Surprises Aren’t Surprising

Robert Gottliebsen (The Australian 22 June) stated that “last week Telstra experienced a sudden profit downgrade. This week Flight West collapsed. Just like their counterparts in London and New York, Australian investors will need to get used to surprises in coming months.” Rather than accept these ‘surprises’ in a meek and sheep-like manner, Australians might instead read David Dreman’s article “Nasty Surprises” in the 19 July 1993 issue of Forbes (an extended, elaborated version appeared in Financial Analysts Journal, vol. 51, July-August 1995) or Chap. 6 of his book Contrarian Investment Strategies: The Next Generation (1998). If they do so they will find that Dreman has analysed almost 100,000 analysts’ forecasts and assessed their correspondence to reality. This enables Dreman to place financial ‘surprises’ on a systematic and statistical rather than a flimsy and anecdotal basis. 

According to Dreman, so-called earnings ‘surprises’ “have an enormous, predictable and systematic influence on stock prices”. What Table 1 implies, Dreman substantiates: analysts systematically and drastically over-estimate both the extent to which they are able to forecast companies’ earnings and the magnitude of these earnings. Analysts, in other words, not only err greatly; the spectacles which blur their vision are deeply rose-coloured. Accordingly (and to the extent that they systematically estimate value), analysts over-estimate favoured stocks’ value and thereby encourage individuals and institutions to pay through the nose for ‘growth.’ A trigger such as an earnings surprise subsequently causes analysts precipitously and drastically to revise their assessments of their favourite stocks; this re-assessment, in turn, causes their market capitalisations (and perceptions of ‘growth’) to tumble. “Consider the ‘best’ companies. Investors, fortified by. brokers and advisors, expect only glowing prospects for these stocks. After all, analysts confidently – overconfidently – believe they can divine the future of a ‘good’ stock with precision. These stocks are not supposed to disappoint; people pay top dollar for them for exactly this reason. So when the negative surprise arrives, the results are devastating.”

Don’t They Ever Learn?

Dreman (and, not incidentally, a growing literature in the field of behavioural economics) thus subsumes under more general principles market participants’ predisposition to err on the side of exuberance. He therefore describes the “psychology of misjudgement” (Charles Munger’s phrase) which occurs repeatedly in financial markets. In this context another of Dreman’s results is even more disconcerting: “strong psychological forces compel investors to buy sizzling issues, and then prevent them from even analyzing where they went wrong” (italics added). Market participants, in other words, not only err systematically and repeatedly: many (like the member of Mensa profiled in Letter 18) seem to be incapable of identifying and learning from their mistakes.

Telstra’s is most prominent, but one need not look far (Australian Gas & Light, Coles-Myer and Lend Lease come immediately to mind) to find other recent Australian examples of Dreman’s earnings surprise syndrome. One can therefore think independently and thereby guard oneself against surprises which might occur at some point in the future. On 19 June, for example, this country’s most popular television program devoted to personal finance and investment commended to its viewers’ attention three companies listed on the ASX. Each is strongly recommended by ten of Australia’s most prominent brokers. With respect to the first company, 4 regard it as a ‘strong buy,’ 2 as a ‘moderate buy’ and 2 as a ‘hold’ (two of the ten have not analysed it and none of the ten rate it as a ‘sell’). Three brokers rate the second company as a ‘strong buy,’ 2 as a ‘moderate buy’ and 5 as a ‘hold’; accordingly, none say ‘sell.’ With respect to the third company, 1 broker regards it as a ‘strong buy’ and 4 say ‘hold’; five have not analysed it and (surprise, surprise) none rate it as a ‘sell.’

At the close of trade on 19 June, these companies’ shares sold for an average of 55.5 times their combined earnings during the two previous six-month reporting periods. In terms of the price paid for their earnings per share, these companies are almost exactly three times dearer than Telstra. But the TV host forgot to mention the heroic assumptions which must accompany their purchase at these prices. Indeed, if these companies are simply to generate the same (to say nothing of greater) ‘coupons’ as those guaranteed by a ‘risk-free’ five-year Commonwealth Government bond, then during the next five years their E.P.S. must grow at an average annual compound rate of no less than 74%. Terra Australis, the Great Southern Land, contains many flora and fauna which amazed explorers and enthral today’s tourists; alas, it contains few if any enterprises which can sustain such rates of growth for any period of time. 

Sobering Implication No. 2 If Dreman’s earnings surprise syndrome applies to Australia as much as it does to the U.S., then not only are further ‘surprises’ waiting to happen: Australian market participants, imprisoned within the walls of EMH dogma, are largely incapable of identifying the causes of ‘surprises’–and hence of learning from their repeated mistakes. 

An Appeal to Valuation and Caveat Emptor

In sharp contrast to the vast majority of Australian financial academics, advisors, analysts, brokers, funds managers and journalists, Grahamite value investors eschew advanced mathematics, warm expectations, passing fads and institutional imperatives. Instead, they embrace hard evidence, cold and straightforward logic, enduring principles and Emersonian self-reliance (“Whoso would be a man must be a nonconformist. He who would gather immortal palms must not be hindered by the name of goodness, but must explore if it be goodness. Nothing is at last sacred but the integrity of your own mind. Absolve you to yourself, and you shall have the suffrage of the world.”)

Focussing their attention both upon investment fundamentals and history, Grahamites thereby inoculated themselves against the ebullience, over-confidence, overvaluation and proclivity to misjudge which since the late-1990s have characterised mainstream views about Australian blue chip stocks. Analysing matters at a distance from the fray, like the Impartial Spectator in Adam Smith’s The Theory of Moral Sentiments, they also subsumed dispassionately the events following Telstra’s 12 June announcement under general rules of thumb. During the second half of June there occurred a flurry of writing and commentary about Telstra by financial journalists and analysts. From this outburst of words (together with gems such as the report entitled “Great Start: One.Tel Is Here To Stay” issued in April by Goldman Sachs) emerges an important insight. If (as Mr Buffett once stated) “forecasts may tell you a great deal about the forecaster; [but] they tell you nothing about the future”, then analyses and reports may tell you much about analysts and journalists – but little or nothing of value about the subjects of their analyses (or finance and investing more generally). 

In the wake of its announcement to the ASX, remarkably little of any significance was written or broadcast about Telstra; volumes, however, could be discerned about the advisors, analysts, brokers, funds managers and journalists who made these statements. Although there are honourable exceptions (Alan Kohler’s articles in The Australian Financial Review of 19 June and The AFR Weekend Edition of 23-24 June are excellent examples), responses to the announcement seemed at first glance to be knowledgeable but upon examination were often vacuous. (Indeed, more than one brought to mind Benjamin Franklin’s aphorism that “he was so learned that he could name a horse in nine languages; so ignorant that he bought a cow to ride on”). These comments and reports tended to consist in idle assertions rather than conclusions derived explicitly from plausible premises and valid logic; convictions stated with confidence rather than inferences expressed with caution; malapropisms and bastardisations rather than justifiable applications of academic research; and feeble excuses rather than robust mea culpas

Hence according to Alan Kohler “the professionals are obviously no wiser to what is going on than [private investors]”. Their possession of apparent rather than real expertise can cause great and widespread harm. According to a four-year study summarised in The Wall Street Journal on 12 June, “investors lost an average of 53.3% when they followed the advice of an analyst employed by a Wall Street firm that had led or co-managed a particular stock’s initial public offering of stock. By contrast, investors lost just 4.2% when they took the suggestions of analysts whose firms had no underwriting relationship with the companies researched”. The WSJ alluded to but did not emphasise the study’s most startling conclusion: to follow analysts during these years was to lose money. To be sure, some kinds of analysts generated greater losses than others; on average, however, all did so. 

Sobering Implication No. 3: Although it is likely unconscious and unintentional, analysts’ modus operandi does not completely exclude the dissemination of sunny, woolly, egregious and self-serving nonsense. Accordingly, and as Thomas Donlan (Barron’s 18 June 2001) editorialised, “a person who buys stock on the unsupported word of an analyst is not an investor; he’s a sucker.”

Chris Leithner


Contact | Disclaimer | Subscription
Chris © 2014-2017 by Artist Web Design