You must predict the future, yet the future is not reliably predictable.

The difficulty of predicting the future even a few years ahead. 

An unresolvable contradiction exists: to perform present value analysis, you must predict the future, yet the future is not reliably predictable. 

The miserable failure in 1990 of highly leveraged companies such as Southland Corporation and Interco, Inc., to meet their own allegedly reasonable projections made just a few years earlier-in both cases underperforming by more than 50 percent-highlights the difficulty of predicting the future even a few years ahead. 

Investors are often overly optimistic in their assessment of the future. 

A good example of this is the common response to corporate write-offs. This accounting practice enables a company at its sole discretion to clean house, instantaneously ridding itself of underperforming assets, uncollectible receivables, bad loans, and the costs incurred in any corporate restructuring accompanying the write-off. 

Typically such moves are enthusiastically greeted by Wall Street analysts and investors alike; post-write-off the company generally reports a higher return on equity and better profit margins. Such improved results are then projected into the future, justifying a higher stock market valuation. 

Investors, however, should not so generously allow the slate to be wiped clean. When historical mistakes are erased, it is too easy to view the past as error free. It is then only a small additional step to project this error-free past forward into the future, making the improbable forecast that no currently profitable operation will go sour and that no poor investments will ever again be made. 

How do value investors deal with the analytical necessity to predict the unpredictable? 

The only answer is conservatism. 

Since all projections are subject to error, optimistic ones tend to place investors on a precarious limb. Virtually everything must go right, or losses may be sustained. 

Conservative forecasts can be more easily met or even exceeded. 

Investors are well advised to make only conservative projections and then invest only at a substantial discount from the valuations derived therefrom.