Stock prices move up and down every minute due to fluctuations in market sentiment, supply and demand. So many factors can affect the stock prices over the short term. With this research we want to know what parameters are most correlated to stock price performances over the long term. We studied the effects of earnings per share, revenue per share, free cash flow per share, EBITDA per share, operating income per share, pre-tax income per share, book value per share and tangible book value per share on the performances of stock prices. According to earlier research, we believe the best way to do this research is to check the performance of business and stock prices over complete market cycles.

We think market peaks appeared at 1999 (we use 2000 in this case due to data availability), 2007 and 2012, while market troughs appear at 2002 and 2008. Based on this, we research the stock price changes from peak-to-peak (2000 to 2007, 2007 to 2012 and 2000 to 2012) years and trough-to-trough years (2002 to 2008). Since financial companies are capital-intensive business with plenty of assets on the books, we separate the companies into two parts: financial companies and non-financial companies.

Of course, 2012 might not be a market peak. But it should be much closer to market peak than market bottom.**Methodology**

1. Calculate the stock price change ratio for four periods, including 2000 to 2007, 2007 to 2012, 2000 to 2012 and 2002 to 2008. Take logarithm of the price change ratio to base 10.

2. Calculate the business performance parameters change ratios for four periods and then take logarithm to base 10. These parameters include earnings per share, revenue per share, free cash flow per share, EBITDA per share, operating income per share, pre-tax income per share, book value per share and tangible book value per share.

3. Do regression of stock price change ratio and business performance ratio for each period and then find which parameters are most correlated to stock price changes.**Results**

We separate companies into two categories, financial and non-financial, to see the difference in different industries. They are separated in this way because financial companies tend to make money from assets, and non-financials make money from operations.**Financial Companies**

In statistics, the coefficient of determination, denoted R square, indicates how well data points fit a line or curve. In our case, we use the linear regression. The higher the R square and adjusted R square, the better the stock price change can be explained by the factor change. From the above four tables, we can see for financial companies, stock prices changes are most correlated to book value changes and tangible book value changes.

From the above chart, we can observe that the log (price in 2012 / price in 2007) can be explained by log (book value in 2012 / book value in 2007) by 61%.

To make our research more complete, we want to see how the price to book value changed in these four periods. The horizontal axis is the price to book value in beginning year of each studied period. The vertical axis is the ratio of P/B in ending year over P/B in beginning year of each period. If the ratio is 1, it means that the P/B ratio for the stock did not change. If the ratio is bigger than 1, the P/B ratio of the stock expanded. Therefore, the stock prices may have benefited from both the increase of book value and the expansion of P/B ratio over time. If the ratio is smaller than 1, it means that the market has assigned a lower valuation for the stock over time.

The probability of P/B, price, and book value expansion are listed below, as well as the median and average of price expansion and book value expansion:

We rank the initial P/B ratios from the lowest to highest and separate them into four parts. The first interval represents the 0 to 25th percentile of total amount of initial P/B ratios. The second represents the 25th to 50th percentile. The third refers to the 50th to 75th percentile, and the fourth indicates the 75th to 100th percentile. From the above charts and table, we can see for the higher P/B ratio at the beginning of the studied periods, is the more likely it will decline, and vice versa. If the P/B ratio at the beginning of the periods is less than around 1, the P/B has a high chance of expansion. If the P/B ratio is higher than around 2, the chance of P/B expansion is much smaller. Most likely the P/B will decline. The book value has similar expansion no matter what initial P/B ratio is. Therefore, the lower the initial P/B ratio, the higher chance of P/B expansion will be.

One concern investors might have is why the market assigned lower P/B ratios for some companies while assigning higher P/B ratios for others. Do companies that have lower P/B ratios initially have operational problems? To get some answer for this question, we looked at the growth of the book value over the period. The results are shown in the last three columns of the above table. We found that the growth of the book value for the companies with lower initial P/B might be slightly lower than the book value growth for those with higher initial P/B. But it still pays to buy stocks with lower P/B, because over a complete market cycle the expansion of the P/B ratio more than compensates for the relatively lower growth rate, resulting in much higher investment returns.**Non-Financial Companies**

From the above four tables, we can see for non-financial companies, stock prices changes are most correlated to book value changes, EBITDA changes and operating income changes. Though they only explained around 50% of the changes in price from 2000 to 2012.

Here we want to see how the P/B, P/EBITDA and P/EBIT (we use EBIT to represent operating income) change in these four periods. Similarly, the horizontal axis is these ratios at the beginning of each period. The vertical axis is the expansion ratio of these parameters at the end of each period.

The P/B ratios for non-financial companies are generally higher than those of financial companies. From 2007 to 2012, the P/B ratio of most companies did not change as much as in the other periods. In most cases, the P/B ratio increased a lot compared to the beginning year. From table 4, we can also observe the same result as from table 2 that the good investing strategy could be to invest in stocks with a lower P/B ratio.

From 2007 to 2012, the P/EBITDA ratio of most companies did not change a lot, while in other periods, the extreme changes occurred more often. From 2002 to 2008, the P/EBITDA ratio of majority companies declined. Similar to P/B ratio, if the P/EBITDA ratio at the beginning of the periods is less than around 3.5, the P/EBITDA has a high chance of expansion. If the P/B ratio is higher than around 10, the chance of P/EBITDA expansion is much smaller.

Similar to financial companies, the companies with higher initial P/B, P/EBITDA and P/EDIT tend to have higher growth rate over the same period. But over time, it still pays out to hold relatively cheap stocks. The expansion from P/B, P/EBITDA and P/EBIT exceeds the higher growths of more expensive companies. The strategy of buying cheaper stocks still pays off. **Conclusion**

From the above research, we can see that for financial companies, stock price change is more correlated to the change of book value during market peak to peak. For non-financial companies, stock price change is more correlated to the change of book value, EBITDA and operating income from market peak to peak. And during market trough to trough, the correlation is little. This is probably because during that period, the stock price would be hit by extreme market changes, government intervention and so forth. Investor expectation would also affect the stock price and company valuation cannot be a good indicator to value the stock.

Interestingly, one important parameter, free cash flow, does not have strong correction with the stock price. This is a surprise we had from the research. Is free cash flow overrated for stock valuation? We will discuss this in future articles.

Another surprise we had was about P/B. Often ignored these days, P/B is actually the parameter that has the most correlation with the stock performance.

The winning strategy we conclude from this research is to buy stocks with low prices to EBITDA or low prices to book. Stocks with lower prices to EBITDA or low prices to book have much higher chance for ratio expansions. Investors can benefit from both the growth of the business and the expansion of the ratios.

This All-In-One Screener link will allow you to screen stocks whose P/Bs are at the bottom 10% percentile within their industry. You can add other filters to it and find the stocks with the most upside potential.

[img]null[/img]

AlbertaSunwapta- 2 years ago