If price/earnings is the proper way to value stocks, then stocks are cheaper at this moment than they have been in more than twenty years. Here's a historical comparison and some thoughts on what it means.

The price/earnings ratio is generally considered to be the best measure of whether stocks are "cheap" or "expensive." There are, however, several ways to measure price/earnings ratios.

The first major distinction in PE ratios is the difference between trailing PEs and forward PEs.

Trailing PEs measure the price of a stock against the past twelve months of earnings. This measurement tends to base the "value" of a stock on the trends it has shown in the past, with the assumption that those trends are likely to continue.

Forward PEs take the current price of the stock and divide it by the estimated earnings in the coming twelve months. The forward PE bases the value of a stock on the expectations for the growth of the stock, regardless of trends in the past.

In general, forward PEs can be viewed as how the market values the future, whether based on speculative expectations or projections of proven fundamental trends.

The argument for using forward PEs, instead of trailing PEs, for evaluating a stock is that the future is what the market pays for, not the past.

One problem with analyzing forward PEs over a great length of time, however, is the availability of data. Estimates for earnings change frequently, particularly after a quarterly report or altered guidance.

Should the price to be used for a forward PE be the price on the day that actual trailing earnings are published? This might be misleading, since changes to estimates generally occur one to two weeks after a quarterly report or guidance, and not all analysts publish their changed estimates on the same day.

Trailing PEs, on the other hand, use the fully reported actual earnings and the price on the day that the PE is measured.

Another argument for using trailing PEs instead of forward PEs is that many earnings estimates are not true reflections of expectations for a company.

Few analysts will forecast an earnings estimate that is higher than guidance given by the company, even when the analyst and the market fully expects the company to beat the guidance numbers. This means that the forward PE for that stock is artificially high since the denominator in the ratio is artificially low.

While forward PEs are useful for individual stocks, in general, when stocks as a group are evaluated, trailing PEs are most frequently used.

The reason for this is that, over time, the overall expectations for groups of stocks tends to even out in a group, with excessive expectations for some stocks balanced by overly pessimistic expectations for others.

In addition, the pricing of a stock one year later tends to reflect how well the expectations for the forward PEs of a year earlier were met.

The difference between operating earnings and as-reported GAAP earnings is that one-time, non-recurring charges are excluded from operating earnings.

The market tends to focus on operating earnings when valuing stocks. Over time, the impact of one-time charges does affect the health and growth of a business, but the short-term focus of many traders tends to overlook any major difference between operating earnings and as-reported GAAP earnings.

With all of this in mind, we decided to analyze the value of stocks today, compared to how the market has valued stocks over the past twenty years.

The following chart illustrates the trailing PE from operating earnings of the S&P 500 Index over the past twenty-three years. We choose operating earnings for this chart because of the extreme aberrations of the fourth quarter of 2008. The current trailing PE on the S&P 500 is 12.4.

Source for data: Standard & Poor's; chart by Briefing.com

It should be noted that the PE for the six quarters between now and December 2012 are based on current operating earnings consensus estimates, as used by Standard & Poor's. The most recently reported quarter, ended June 30, 2011, is listed as an actual, although there is still 1% of the S&P 500 Index that have not reported their 2008Q2 actual earnings.

The first single and direct conclusion to be drawn from this chart is this:

Stocks have not been as cheap as they are today since December of 1988.

Whether this means that the price of stocks will increase is a matter of debate, of course.

The downward trend seen in the chart for the upcoming six quarters is calculated using today's price, making the PE for the 2011Q3 through 2012Q4 quarters forward PEs. (These data points are shown in green in the chart above.)

This leads to the ability to project the potential for the S&P 500 in the future, making assumptions about what the trailing PE for the index will be once those projected quarters earnings become actual earnings.

The following table illustrates the potential price of the S&P 500 Index, assuming that the current trialing PE of 12.37 is maintained.

Source for operating earnings estimates: Standard & Poor's

This means that if the current trailing PE were maintained, the overall market would be 16% higher than current levels one year from now.

Will the current trailing PE be maintained? If you make the assumption that the current index level of 1148 is maintained, instead of the PE multiple (even while earnings rise), then the following trailing PEs can be projected for the upcoming quarters.

Source for operating earnings estimates: Standard & Poor's

These PEs, of course, are the same as the forward PEs shown in the chart above.

The concept of whether a stock is "cheap" or not depends primarily upon the comparison of its current price to its future price.

There are two main variables to consider when considering the possible future value of stocks.

• Projected earnings -- whether the projected increase occurs or not
• Projected multiple -- whether the current multiple upon earnings remains the same or changes

This combination of possibilities leads to four different scenarios.

If you make the assumption that the current trailing PE will stay flat for the near future and that the projected earnings are accurate, then the S&P 500 has the potential to be 20% higher at the end of 2012 than it is today, as illustrated above.

This assumption of the trailing PE holding at current levels, however, is contradictory to the current trends of continually declining PEs, as shown in the table above.

If you make the assumption that the current PE multiple stays the same, but that earnings do not increase, then the index would be flat. However, if the projected earnings do not increase, it is hard to see why the market would maintain current multiples on earnings.

A scenario of falling or flat earnings combined with rising trailing PEs is hard to envision.

On the other hand, if you make the assumption that the trailing PE will rise towards its twenty-year average of 19.0, and that the projected earnings will indeed occur, then the potential increase in the S&P 500 Index is far greater than 20%. The combination of rising earnings and multiples placed upon those earnings results in extremely high returns.

To answer the question of "are stocks cheap?" requires making a decision about these two variables.

The ironic situation in the market today is that earnings have actually been increasing, at consistent and strong rates, while the multiple on earnings has sharply declined.

At some point, if you believe that the price of stocks is ultimately based upon the potential for earnings growth, than this combination of rising earnings and declining multiples cannot last. Such a belief implies strong potential growth in the price of stocks in the future, which can also be phrased as "stocks are cheap."

On the other hand, if you do not believe that the projected earnings will in fact occur, then it is likely that the trend of declining multiples on earnings will continue, the combination of which would lead to much lower levels of stock prices. With such a point of view, the current level of stock prices is probably overvalued.

Is it reasonable to believe that the projected earnings for stocks in general will continue to rise, as they have steadily since the fourth quarter of 2008?

On this question rests the entire issue of whether stocks are "cheap" or not.

We tend to believe that the projected earnings over the next six quarters will, in fact, turn out to be accurate, at least in the overall trend.

Is it likely that the overall market will continue to place no value on such rising earnings and maintain indexes at current levels, driving trailing PEs ratios to all-time lows? That question is harder to answer.

After all, during the Internet bubble, we frequently stated that "just because stocks are overpriced does not mean they can't go higher." Today, it seems equally reasonable to state, "just because stocks are very cheap does not mean they can't go lower."

If, of course, you do not believe that the value of stocks is based in any way upon trends in earnings, then there is little reason to be in the stock market in the first place.