HOME

  

  

Recent Press Comments on STAR

Download press comments on STAR in pdf format

 

October 25, 2003, Financial Times - Saturday Surveys

STOCK-PICKING

A 'mechanical' approach

Philip Coggan continues his appraisal of selection methods. This month: the STAR share system

After two months looking at how professional fund managers pick stocks, it is time for an examination of another
"mechanical" share selection system.
The star share system was set up by John Mulligan, who worked in the City before joining the United Nations as a consultant economist. He set up a financial advisory company, MD Management, in 1986 and spent a long time researching the relationship between earnings' forecasts and the subsequent performance of share prices.
Having discovered a link, he set up a newsletter more than a decade ago and has been making recommendations since then. He has back-tested the portfolio to 1986 and has found quite substantial outperformance, with the star system beating the FTSE All-Share index in capital gains terms 13 times in 17 years.
The star system managed to make money in both 2000 and 2001, although it did fall 16.4 per cent in 2002. He calculates that someone who invested 1,000 in the star system at the end of 1985 would have accumulated 17,620, before costs, by the end of last year, compared with just 2,660 had they invested in the All-Share.
Mr Mulligan's starting point is to calculate the prospective price-earnings ratio (based on earnings two years' hence) for his universe of 300 stocks. Having found the top quartile (75 stocks) of that group, he then selects the companies with the fastest earnings growth. He winnows the list into two portfolios, one with 10 stocks, one with 20.
He uses two sieves before the list is complete. First, he eliminates any company where earnings forecasts have recently been downgraded by more than 20 per cent. All too often, such downgrades can be the start of a long and slippery slope.
Second, to control sector risk, he ensures that there are no more than two companies from an individual sector in the 10-stock portfolio and three in the 20-stock portfolio.
The criteria tends to select growth stocks but for those who need income, Mr Mulligan revises the process one stage further, picking the stocks from top quartile with above median dividend yields.
His back-testing has shown this portfolio has also outperformed the All-Share in capital terms.
Before we reveal the stocks selected, a reminder of the ground rules. First, any system that selects a small number of stocks is bound to be risky. Second, there is a gap between the writing of this column and its publication - bad news may have occurred in the intervening period.
Third, this column is definitely not a tip for each individual share; these systems only work if
a whole portfolio is acquired so that the winners outweigh the inevitable disasters. And finally, I do not, of course, own any of these shares myself.
The 10 stocks selected by Mr Mulligan's growth approach on October 7 were Avon Rubber, Brammer, Charter, HBOS, Hiscox, Imperial Tobacco, Lloyds TSB, Mowlem, SVB Holdings, and UK Coal. More details are in
the table. We shall monitor its success over the coming months.

STAR Portfolio as at 7th October 2003

Company

Price (p)

Yield %

Avon Rubber

198

3.8

Brammer

142

3.2

Charter

87

nil

HBOS

716

4.1

Hiscox

149

2.4

Imperial Tobacco

986

3.4

Lloyds TSB

436

7.8

Mowlem

205

3.6

SVB Holdings

58

0.4

UK Coal

130

7.7

 


Financial Times: Jan 17, 2004 FT MONEY - INVESTING:

Analysts will be put to the test
By Philip Coggan

Every chief executive knows the importance of matching analysts' profit forecasts.
Meet, or slightly beat, expectations, and the company's shares will continue their upward path.
But fall short of those forecasts and your share price will take a beating. Hell hath no fury like an analyst scorned. Once a company gets a reputation for disappointing the market, it can be years before the share price recovers. Usually, a rebound requires a change in management.

The rewards and penalties for meeting analysts' estimates are so great that companies have clearly been tempted to use the full limit of accounting flexibility - and beyond - in order to get the numbers right. Some of the biggest financial scandals of the past few years have flowed from the willingness of companies to massage their numbers in order to meet market expectations.

Furthermore, executives play their part in setting those expectations. Their aim in life is to persuade analysts to make forecasts that show a reasonable level of profits growth (so investors find the shares attractive) but for those forecasts slightly to underestimate the true numbers. That way, the company can produce a positive surprise when the figures are finally published.

In recent months, the trend of earnings estimates has been moving sharply higher. That has helped fuel the stock market's remarkable rally. Many commentators believe the trend in earnings estimates is a vital indicator of the health of the market.

In collaboration with John Mulligan of M D Management, we are setting out to find the answers to two important questions: how accurate are the forecasts?; and is it possible to use the forecasts to pick stocks that outperform the market?

Mulligan accordingly compiled forecasts for a wide range of companies from 10 investment banks and brokers.

The table gives a flavour of their views. It focuses on 10 well known stocks, showing the last recorded earnings per share and the analysts' forecasts for the next two financial years. (All the brokers did not have forecasts for all the companies.)We will be testing to see exactly how close these initial estimates are to the actual outcome, giving points depending on the degree of accuracy. We will also be reporting back every quarter on how the individual banks and brokers have been performing.

Like all forecasters, analysts tend to get their estimates wrong. Sometimes this means that they are too pessimistic about the outlook for profits, usually when an economy is emerging from recession.

Far more often, they are too optimistic. There are several potential reasons for their over-enthusiasm. During the 1990s, it was alleged that many analysts deliberately published upbeat forecasts for company profits in order that their colleagues in the corporate finance department could win business.

The temptation was great. Corporate finance departments found it easier to attract clients if their analysts were optimistic about a company's profits and share price. And once that business was acquired, it was easier to sell securities to investors if the analyst took a bullish view. The motiva tion was increased at banks where much of an analyst's bonus was paid by the corporate finance department.

The pressure was also strong from the other direction. Many executives have remarkably thin skins about the way analysts view their shares. If the company employs a broker or investment bank as an adviser, they expect that firm's analysts to be bullish. A bearish note may lead to a parting of the ways between bank and client. Even if the analyst is entirely independent, the company can make his life difficult by refusing to return calls or excluding him from trips to the group's operations.

Finally, analysts may well be simply upbeat by nature, or become so as a result of their job. The analyst has to compete for the attention of the sales staff at his firm, of his bosses and of investor clients; being optimistic is a better story to sell. After all, if the industry prospers, so will the analyst. And as time passes, it may be harder to be negative about a company, whose executives the analyst has met many times; it is possible to be "captured" by the industry.

This perpetual optimism usually leads to a familiar pattern. At the start of the year, analysts are looking for double-digit profits growth across the market; their strategist colleagues, who look at broader economic trends, are much less sanguine. As the year progresses, analysts gradually lower their estimates until they are closer to the strategists' numbers.

Even if there are systematic biases in analysts' forecasts, it may still be possible to profit from what they say. Many investors, rather than look at the detail of each forecast, look at the trend in figures from the whole analyst community.

That approach can be successful in the short term. It seems as if the market can be slow to react to news. One analyst may adjust his forecasts higher to reflect an improvement in a company's fortunes. But the rest of his colleagues may take weeks or months to notice the same development. As more and more catch on to the change, forecasts will drift upwards and so will the share price. Investors can take advantage of this trend.

Mulligan's star system (featured last year in the FT Money Guide) starts with a group of stocks. He selects the quartile (25 per cent) from that group with the lowest prospective price-earnings ratios. Of those, he selects the companies with the fastest earnings growth.

In constructing a portfolio, he applies two sieves to this list. He eliminates any company where earnings forecasts have recently been downgraded by more than 20 per cent. And he limits sector exposure, ensuring that no more than two stocks from any industry appear in a 10-stock portfolio. He will be applying this approach to the forecasts made by the 10 banks and brokers and we will report back on how the chosen stocks perform.


Stock Picking: Now, for a progress report By Philip Coggan
FT Your Money; Feb 25, 2004

It is time to give an update on the progress of the various stock-picking portfolios this series has been examining during the past 18 months.

All the systems were designed to give the investors scope to pick stocks that could outperform the indices through the application of simple, mathematically-based rules. The idea was to find stocks that looked undervalued on the basis of measures such as dividend yields or earnings growth.

The star portfolio was chosen in October through a method devised by John Mulligan of MD Management. Mulligan tracks earnings forecasts for a universe of 300 stocks and selects the quartile with the lowest prospective price-earnings ratios. Of those, he selects the companies with the fastest earnings growth.

The stocks are winnowed into lists of 10 or 20, by eliminating any company where earnings forecasts have recently been downgraded by more than 20 per cent and controlling the sectoral representation to just two companies in a 10-stock portfolio (and three in a 20-stock portfolio).

October's selection process produced the following 10 stocks: Avon Rubber, Brammer, Charter, HBOS, Hiscox, Imperial Tobacco, Lloyds TSB, Mowlem, SVB Holdings and UK Coal. As at the end of January, that list had modestly outperformed the FTSE All-Share index, gaining an average 4.2 per cent against the index's 3.8 per cent rise.

In the same month, we updated the O'Higgins portfolio of high yield stocks. The O'Higgins approach was honed using the Dow Jones Industrial Average in the US; it selects the 10 stocks with the highest dividend yield and, of those, chooses the five with the lowest share prices.

In the UK, we apply the system to the now little-used FT 30. This worked well in the 1980s but the portfolio chosen in 2002 (the first in the series) was not successful. Nor is there any good news about the five stocks chosen in mid-October: BAE Systems, EMI, P&O, Royal & SunAlliance and Tate & Lyle. That portfolio was down 2.3 per cent by the end of January.

Another system that was updated in October was the asset-backed approach. This simply buys the stocks with the lowest price-to-book (or asset value) found in Company Refs (published by HS Financial).

In October, the system picked 10 stocks, but one (Queens' Moat Houses) should have been excluded since its shares were already suspended. The remaining nine (Royal & SunAlliance, Mitchells & Butlers, Land Securities, Liberty International, British Land, Corus, London Merchant, Minerva and Goshawk Insurance) have done well, returning an average 11.8 per cent, compared with a 1.8 per cent rise in the All-Share during the same period.

Looking back further in the year, in June we featured a system that selected stocks based on analysts' earnings forecasts. The 10 stocks chosen were mmO, Shell, Vodafone, BSkyB, SAB Miller, BG, Man, British Land, BP and GUS.

This portfolio has also performed well, rising 11.8 per cent against an 8 per cent advance in the All-Share.

The best, however, has been saved till last. The Caley portfolio was selected on the basis of methods described by stockbroker Raymond Caley in his book, "How to choose stockmarket winners" (published by Piatkus Books).

To qualify under Caley's system, stocks must have three characteristics.

First, they must make a new high over 12 months or longer.

Second, the company's profits must be forecast to break previous records.

Third, the stock's price-earnings ratio must be at a 25 per cent discount to the sector average.

Only four shares met all those criteria last May: Bloomsbury, Martin International, McCarthy & Stone and Telecom Plus. Each of the four has risen by at least 20 per cent since then and the average gain has been 48.4 per cent, 35 percentage points ahead of the All-Share. So far, the system has been a triumph but it is, of course, a test over just one period.

We will look to update Caley's approach (and one or two of the other systems) in May. In the meantime, we will go back to looking at the stock-picking approaches of the professionals in the next two issues.

The ground rules for this series have been as follows. First, any system that selects a small number of stocks is bound to be risky. Second, there is a gap between the writing of the columns and publication - bad news can occur in the intervening period.

Third, the columns are not tips for individual shares; these systems only work if the whole portfolio is acquired so that the winners outweigh the inevitable losers. And fourth, I have not, of course, owned any of these shares myself.

 

 

 

 

 


Copyright MD Management Limited 23004. Reproduction in any format is prohibited without prior authorisation from MD Management Limited who cannot be held liable for any inaccuracies in content.The information on this website is issued by MD Management and is not a personal investment recommendation. The information was obtained from sources believed to be reliable; but MD Management does not guarantee its accuracy or completeness. You should be aware that the STAR methods use objectively determined value investing criteria to rank leading shares in each period monitored and there can be no guarantee that past performance over any period will be repeated in the future. The STAR share selection methods do not take any account of dealing costs, margins or taxation and investors using the STAR ranking lists are recommended to seek professional advice regarding the suitability for you of any investment that you are contemplating. MD Management does not offer any investment advice. We accept no liability for any direct or indirect or consequential loss arising from the use of this document. However, nothing in this provision shall exclude or restrict any obligation owed to you under the Financial Services Act 1986, or the Rules of the Financial Services Authority.