Earnings get all the attention on Wall Street. While earnings are of course important (and measures like P/E are very useful in stock selection), they aren’t everything. Less discussed, but equally important, is the cash that is flowing into and out of a company. Hidden in the statement of cash flows is an extremely useful tool for identifying great investments.
When a company releases its three financial statements, the income statement (led by bottom line earnings) dominates, the balance sheet comes second, and the statement of cash flows generally comes third. You’ll see a lot of “breaking earnings reports” on CNBC, but you’ll never see a “breaking cash flow report.”
This may be in part because the statement of cash flows has only been around since 1987, whereas income statements and balance sheets have been around forever. In the cash flow statement, cash flows are grouped into operating, investing and financing categories. Operating cash flows dominate—these are inflows and outflows resulting from normal business operations. Investing cash flows matter too, because they reveal how much a company is investing in things like machines, office buildings, and the like. But financing cash flows are the most useful for investors.
Financing cash flows are pretty straightforward. Positive numbers (cash coming into the company) result when a company raises cash from outside stakeholders (through debt or equity offerings). Negative numbers (cash leaving the company) result when a company sends cash back to stakeholders by repaying debt, paying dividends, or buying back shares.
It turns out that companies with the most negative flows, relative to their market value, perform considerably better than those with the most positive flows and perform better than the overall market. The factor (financing cash flows divided by market value) is sometimes called “shareholder yield,” but I prefer “stakeholder yield” because it doesn’t belittle the creditors who are not technically equity shareholders. Here are the annualized results, since 1987, for stocks broken into decile groups by stakeholder yield.
The bottom line is that following financing cash flows has been extremely useful in the past, and you ignore them at your own peril.
NOTE: This factor has received attention (notably in Jim O’Shaughnessy’s What Works on Wall Street, in Meb Faber’s Shareholder Yield, and in academic papers), but still remains fairly under the radar. In this example, I only run the numbers since the statement of cash flows was first reported in 1987 and only use data directly from the cash flow statement (although the factor can be extended to 1970 using other data, as I did in a previous piece).
/rating_off.png "1 Star")
Loading...
excellent post, keep up the great work. can you provide and more granularity on the data? links to your work/academic papers you referenced. thanks!
Very useful indeed. However could you give some more details about the backtest -
1. What is the universe of stocks used?
2. Subperiod-performances (or may be use a log-scale graph?)
3. Most importantly, which fundamental data-set did you use and did you lag the cash flow data to take care of data availability issues, and if yes then how many months lag?
1) stocks in the U.S. with market caps above $200MM (inflation adjusted back in time)
2) can’t do log scale with negative values but I can do a follow up post with base rates
3)Cash flow data is lagged 3 months
Thanks! Keep posting.
very nice article. We did a research piece on this last year too and had similar observations. You may download the piece here.
https://jpmm.com/research/content/GPS-1097463-0
Great Blog. Here is our paper on the subject of shareholder yield. http://papers.ssrn.com/sol3/papers.cfm?abstract_id=2051101
Wes, big fan of yours! I reference Quantitative Value and your website in my upcoming book. Cheers!
Cannot download this report.
Very nice blog with enjoyable and thought-provoking posts! I hope you’d be able to clarify the case a little. What kind of share/stakeholder yields are you seeing over time in the top 1-3 deciles? Just wondering about the range we’re looking at there. Many thanks for your posts and time put into them!
Really interesting analysis, particularly when combined with your "Talk is Cheap" post. You mention in the ending note that you were able to extend the cash-flow data back to 1970 using other data. Would you share that method and point to the data source?