The retail public enjoys the storytelling of growth stocks and the financial press is more than happy to fill that void. It is easy for an investor to fall into what finance professor, Jeremy Siegel, refers to as a "growth trap", focusing on stocks with rapid sales and earnings growth in new industries that are highly disruptive. Humans have a predisposition to see patterns where none exist. Connecting the dots and extrapolating above average growth trends far into the future is an example of this pattern seeking behavior. Let's take a look at the evidence and see if past growth in earnings really does continue into the future for stocks we want to invest in.

My former professor, Dr. Josef Lakonishok, reviewed the performance of stocks that exhibit above average earnings growth over the recent past and found very few of them were able to maintain this stellar earnings trajectory. Over the study period 1951-1997, firms that managed to have a 5 year track record in the upper 25% in terms of past earnings growth, only around 2% of them went on to stay above the median firm growth rate during the next 5 year period.[1] Although the retail public is enamored with growth and is handicapped in seeing growth patterns where none exist, the evidence doesn't seem to show an ability to predict what companies will continue on with above average growth.

Predicting what high growth companies will continue their above average growth trajectory is difficult. Can we tip the odds in our favor by including additional filtering rules to be able to predict what high growth companies deliver above average earnings growth into the future? Unfortunately, additional filters such as size or book-to-market screens applied to high earnings companies still fizzle out in terms of delivering those select stocks that continue on with above average growth into the future. After 4 years, only around 2-3% of high growth companies that are screened based on size or book-to-market continue to stay above the median growth rate in terms of earnings. Practically none of these companies still maintain a sales or earnings growth rate above the median after a decade.[2]

So what if the earnings growth dwindles over time for companies that produced above average growth in the recent past. If they outperform the broad market average, aren't they still worth buying? Not if you use only earnings-per-share growth as your buying criterion. O'Shaughnessy in his book, What Works on Wall Street, showed that a portfolio of 50 stocks with the highest previous 5 year earnings-per-share growth rate lagged the market over a 50 year study period. Even breaking down the long study period into rolling 5 year periods, growth companies still lagged the broad market average most of the time.[3]

So in summary, earnings growth can't be predicted and purchasing historical high growth stocks doesn't seem to outperform the market over the long term. What should investors do in terms of a rationale process that would be a better use of their investment capital. The answer is invest in stocks trading below liquidation value and abandon the earnings forecasting game all together.

By focusing your data mining activity not on the income statement, but on the balance sheet of public companies, a better performance result is delivered. A selection criterion outlined by value investing pioneer, Benjamin Graham, called for purchasing stocks trading below net current asset value, an approximate measure of the liquidation value of a public company. This criterion calls for subtracting all liabilities including preferred stock from the current assets listed on a company's balance sheet. If this measure of liquidation value of a company is significantly greater than the market value, purchase the stock.

Below shows the average performance of stocks trading below 75% of net current asset value over the 1956-2009 study period. No more than 10% was invested in any one stock. If only a few stocks were available that met this stringent value investing criterion, the balance of the portfolio remained in Treasury Bills. As indicated on the graph, even with a portion of the portfolio being handicapped by sitting idle in Treasury Bills, the long term performance is excellent relative to a broad market average.

We know that stocks trading below liquidation value have good historical performance regardless whether earnings are positive or negative. The chart below looks at all stocks trading below 75% of net current asset value and divides them into two groups: ones with positive earnings over the previous year and the others with negative earnings. Both groups perform well, but the ones with earnings losses over the previous year perform even better. This same conclusion is also supported in another study by Tobias Carlisle and his colleagues showing monthly returns of negative earnings companies that trade below 2/3rds of net current asset value outperform ones with positive earnings.[4]

Unfortunately, the superior performance of net current asset value stocks that lost money over the previous year comes with a price. As the chart shows, a smaller percentage of stocks trading below net current asset value with negative earnings increased their share price after a year. Average annual returns are better for NCAV stocks that lost money in the past year, but only 57% of them moved higher in price.

For investors willing to embrace a value investment philosophy and purchase only stocks trading below net current asset value, it is better to focus on the balance sheet of a company rather than earnings summarized on the income statement. Stocks whose most liquid assets on the balance sheet exceed all liabilities manage to outperform the broad market average over the long study period from 1956-2009. Earnings growth is not only difficult to predict in the future, but as a stand alone selection criterion it doesn't add much value in terms of superior portfolio performance. The long term evidence shows that selecting stocks trading below net current asset value is a better use of your time and energy than chasing the next earnings momentum stock touted by the financial press.

[1] Louis K. C. Chan, Jason Karceski and Josef Lakonishok. "The Level and Persistence of Growth Rates." Journal of Finance 58, no. 2 (April 2003): 660-661.

[2] Louis K. C. Chan, Jason Karceski and Josef Lakonishok. "The Level and Persistence of Growth Rates." Journal of Finance 58, no. 2 (April 2003): 656.

[3] James P. O'Shaughnessy, What Works on Wall Street, (New York: McGraw-Hill, 1998), 191-196.

[4] Tobias Carlisle, Sunil Mohanty and Jeffrey Oxman. "Ben Graham's Net Nets: Seventy-Five Years Old and Outperforming." February, 2010. Accessed June 10th, 2014. http://eyquemdotnet.files.wordpress.com/2013/09/benjamin-grahams-net-nets-seventy-five-years-old-and-outperforming-full-tables1.pdf.