QUALIPORT
The Rules Of Long-Term Value

By Maynard Paton (TMFMayn)
January 8, 2004

Many investors rely on broker forecasts for their share picks. Typically, the projections concern earnings and dividends per share and relate to the current year and the year after. However, such predictions can be quite unreliable.

Quality estimates

Let's rewind to the start of 2002. At that time, the consensus earnings and dividend estimates for the sixteen shares on the Qualiport's then watch list looked like this:

Company	EPS forecast current year	EPS forecast year after	DPS forecast current year	DPS forecast year after
Allied Domecq (LSE: ALLD)	35.6	39.3	13.1	14.1
Carpetright (LSE: CPR)	49.1	55.5	30.6	34.4
Emap (LSE: EMA)	40.6	44.5	19.7	20.7
Games Workshop (LSE: GAW)	25.2	30.0	11.4	12.9
Halma (LSE: HLMA)	9.65	10.3	5.28	6.06
Imperial Tobacco (LSE: IMT)	64.7	69.8	31.3	33.8
Johnston Press (LSE: JPR)	22.3	23.0	4.50	4.90
Lloyds TSB (LSE: LLOY)	54.5	58.7	33.7	36.9
London Stock Exchange (LSE: LSE)	17.1	18.9	4.23	5.00
Latchways (LSE: LTC)	17.8	23.4	9.07	9.62
Metal Bulletin (LSE: MTLB)	10.4	12.3	6.00	6.27
PizzaExpress (LSE: PIZ)	46.8	54.4	9.81	11.9
Renishaw (LSE: RSW)	19.6	22.8	15.7	16.9
SSL International (LSE: SSL)	16.5	36.9	12.5	13.2
Ultraframe (LSE: UTF)	21.5	23.8	9.75	10.3
Ulster Television (LSE: UTV)	17.3	17.7	9.20	9.83

Two years on, this is how the actual earnings and dividend figures differed from those estimates (in percentage terms):

Company	EPS forecast current year % difference	EPS forecast year after % difference	DPS forecast current year % difference	DPS forecast year after % difference
Allied Domecq	(8.7)	(14.8)	(0.8)	(0.7)
Carpetright	(5.1)	(7.9)	7.8	7.6
Emap	(7.6)	(17.3)	(1.0)	4.4
Games Workshop	14.7	27.3	14.0	31.8
Halma	(6.0)	(17.6)	0.0	(4.1)
Imperial Tobacco	3.5	26.7	5.6	24.1
Johnston Press	(1.9)	14.6	(2.2)	9.5
Lloyds TSB	(17.6)	(41.4)	0.0	(7.3)
London Stock Exchange	6.4	9.5	(14.9)	(14.0)
Latchways	(70.0)	(18.8)	0.1	(3.2)
Metal Bulletin	1.9	(47.7)	(2.5)	(6.7)
PizzaExpress	(19.7)	*	7.0	*
Renishaw	5.6	(13.6)	1.3	1.2
SSL International	(23.0)	(34.7)	(1.6)	(6.8)
Ultraframe	4.2	(16.4)	5.6	7.8
Ulster Telvision	4.6	4.5	0.0	(2.3)

(*Acquired)

This table summarises the accuracy of the watch list estimates.

Actual v. estimate discrepancy	EPS forecast current year	EPS forecast year after	DPS forecast current year	DPS forecast year after
Above +15%	0	2	0	2
+10% to +15%	1	1	1	0
+5% to +10%	2	1	4	3
0% to 5%	4	1	5	1
-5% to 0%	1	0	5	5
-10% to -5%	4	1	0	3
-15% to -10%	0	2	1	1
Below -15%	4	7	0	0

Forecasting follies

Though it's a small sample, two clear conclusions can be made from the watch list study:

1. Earnings forecasts for the current year proved far more accurate than those for the year after.
2. Dividend forecasts proved far more accurate than those for earnings.

Given the study involved estimates covering 2002 -- a bumper year for company profit warnings -- it's not surprising to find most of the predictions turned out to be over-optimistic. Particular earnings howlers included Latchways, Lloyds TSB and PizzaExpress. In terms of over-pessimism, City analysts did not foresee the recovery at Games Workshop and big 'earnings enhancing' acquisitions at Imperial Tobacco and Johnston Press. On average, actual earnings came in about 8% lower than forecast, though the range was anywhere between 70%(!) under to 27% higher.

Further evidence of 'forecasting follies' comes from research unearthed by David Dreman. In his book Contrarian Investment Strategies: The Next Generation, Dreman outlined the accuracy of 94,000 US consensus earnings estimates made between 1973 and 1996. Amazingly, over 40% of these predictions were more than 15% out:

Actual v. estimate discrepancy	Percentage of estimates
+/- 5% or under	29.4
+/- 10% or under	46.8
+/- 15% or under	58.3

After excluding a handful of large errors that skewed the data, Dreman noted the average profit forecast missed by a mighty 23%.

Golden rules

So, with the future always throwing up 'surprises' and analyst earnings predictions akin to the proverbial motorcycle ashtray, what should the buy and hold investor do? These are the three simple rules to judging long-term value:

1. Concentrate first on business quality predictions: Judging whether a company can maintain its competitive advantage over time will prove more important in the long term than short-term broker predictions.

2. Think in general terms: Don't get bogged down with precise profit forecasts. Simple projections, such as 'a lot higher' or a 'bit higher' over multi-year timescales, will prove more useful.

Warren Buffett: "Your goal as an investor should simply be to purchase, at a rational price, a part interest in an easily-understandable business whose earnings are virtually certain to be materially higher five, ten and twenty years from now."

Philip Fisher: "The investor cannot pinpoint just how much per share a particular company will earn two years from now... [but] he should come pretty close in judging whether a sizeable increase in average earnings is likely to occur a few years from now. But just how much increase, or the exact year in which it will occur, usually involves guessing on enough variables to make precise predictions impossible."

3. Buy with a margin of safety: Nobody's crystal ball is perfect every time. To help minimise the share price downside, look for obvious and immediate value. The Qualiport for instance values businesses on their last reported profits and factors in little or no future growth. Read more | more | more.

And finally...

The last word goes to Dreman: "We have seen that estimates carefully prepared only three months in advance, by well-paid and diligent analysts, are notoriously inaccurate. To complicate matters, many stocks sell not on today's earnings, but on expected earnings years into the future. The analysts' chances of being on the money with their forecasts are not much higher than winning the lottery. Current investment practice seems to demand a precision that is impossible to deliver. Putting money on these estimates means you are making a bet with the odds stacked heavily against you."

More: Order Contrarian Investment Strategies: The Next Generation | Why Dividends Beat Profits

The author owns shares in Carpetright, Games Workshop, Halma, Johnston Press and London Stock Exchange.