The general connection between a company’s profits and its stock price is well established. However, both over the short and long run, that degree of connection is not as robust as investors might hope. Clearly, many factors influence individual stock prices: dividends, revenues, and balance sheets among the obvious. While generally the case, these other metrics do not necessarily move in lock step with profits, particularly over shorter horizons. Still, most studies show that profits are the most influential of these factors on stock prices.
Of course, factors unrelated to individual companies are also at work: interest rates, the overall economy, and any number of risks to the general investment environment. Some professional portfolio managers are, in effect, prohibited from acting on any worries they might have about this last set of factors--the so-called “macro” influences. Through a mandate, they are managing portfolios which are always fully invested, so if the economy cracks and the overall stock market drops, these portfolios could potentially take a hit. Under these circumstances and all others, the only objective for active managers is to perform better than the market. This is the world of relative performance, within just one asset class.
The data show that only a modest fraction of actively managed portfolios (roughly one-quarter) do, in fact, beat the market. So while we know that profits play a large role in stock performance, it turns out to be no easy task to calibrate to this metric. Because the market is relatively efficient, there is the suggestion that merely knowing past earnings performance is not the complete answer to realigning portfolios for future performance.
Before getting to the requirement for accurately forecasting the future, it is important to note that even the past record of earnings has serious challenges. In its lead article one year ago, the Financial Analysts Journal (January-February 2016) published “The Misrepresentation of Earnings.” The research was the result of a survey of 400 chief financial officers, and the findings are sobering. Fully 20% of earnings reports were felt by the respondents to be intentionally distorted and by a significant amount: 10%. In an age where “missing by a penny” causes great shock to the market, this is a serious finding. Further confirming the general nature of the distortions, about two-thirds were in favor of higher earnings.
Public companies are required to issue reports in compliance with “Generally Accepted Accounting Principles” (GAAP). While it sounds as if this standard would allow no wiggle room, independent accountants who certify these results cede fairly large degrees of discretion to management in important areas. As most businesses today have some degree of complexity, it is no surprise that there are innumerable occasions where interpretation of accounting entries can, legitimately, vary widely. Indeed, managements have reasonably great latitude and discretion in their recognition of revenues, expenses and balance sheet items. Most of the time (80%, according to the CFO survey) management does a straight-up job.
Certainly, management is not going to announce they have misrepresented their results. The prescription, from the CFOs themselves, is to rely specifically more heavily on the metric, “cash flow,” rather than earnings.
The fact is that one of the earliest lessons I learned in business was that balance sheets & income statements are fiction, cash flow is reality.
It is interesting that while “double-entry” accounting can be traced back as far as 500 years, the very notion of cash flow was historically offensive to accountants. It was less than 50 years ago that public companies were encouraged to add a statement of cash flow to their income and balance sheets and only about 25 years ago that the today’s version of a separate statement of cash flow was mandated.
A report of revenues for a quarter does not, as it turns out, actually mean that money came into the company; rather, “revenue” is merely the recognition that cash eventually will come in, just not necessarily in the quarter of the report. The same is true of “expenses” and many balance sheet entries. On the other hand, cash flow generally tracks what is happening in a company’s bank account. This activity is, of course, not discretionary.
Cash flow is defined as operating earnings plus depreciation and amortization expenses, minus changes in net working capital. Depreciation and amortization are “expenses” in the sense that they relate to assets previously required, but the value of which decreases over time as they age or decrease by virtue of a previously disclosed schedule. The cash for acquiring those assets has already been used, so no cash is reduced when depreciation is recognized, but earnings are reduced because the entry is treated as an expense in that period. Thus, depreciation and amortization are added back to earnings. Working capital is the difference between “current” assets and “current” liabilities that are scheduled to be realized or paid over the next year. In the meantime, the difference sits in the bank account and therefore changes in net working capital are subtracted from earnings.
Most investors analyzing cash flow add another very important step, i.e., to focus on the future sustainability of the company. Therefore, any capital expenditures made to sustain the company are subtracted from cash flow, even though the accounting “expense” related to the item can be spread out over the active life of that expenditure, even if the cash in the bank went down. Cash flow after capital expenditures is called “free cash flow.”
There are a number of lesser adjustments that can--and often are made--on the statement of cash flow. Together they form what CFOs think, based on the FAJ research, is a more reliable metric of corporate results.
To demonstrate the potential for using the cash flow metric, we ran a very simple test. Looking at the S&P 500, we wanted to know the predictive power of GAAP earnings on the next month’s stock price and compare that with the predictive power of a very simple measure of growth in free cash flow. We then added a third accounting measure to the study, “operating earnings.” The latter, while essentially footnoted to GAAP results, has become the metric most associated with the quarterly frenetic media coverage of earnings.
Essentially, management adjusts GAAP earnings for whatever they believe is a non-recurring event, in order to convey a truer underlying picture of the company. But, to the extent GAAP earnings can reflect some discretion, it does not take a great leap to assume an even greater degree of controlling the earnings image of a company through the use of operating earnings (a trend itself that accelerated over the past two decades).
The charts on the next page show the results of dividing the rate of growth of the three metrics into 30 “fractiles.” The first 10 fractiles are therefore the highest growing third of the universe (in this case, the companies in the S&P 500 Index), and so on. We asked what has happened to the stock price of the companies in each fractile over the next month. The data show the average results per fractile over the past 15 years.
The only reason to start a business is to grow cash. Ultimately, it’s not to create earnings per share or revenue growth. It’s to create free cash flow after all expenses.
The charts show a dotted line which represents the best statistical representation (technically, the line of “best fit” or trend line characterizing the data points) of the relationship between the three ways of expressing corporate accounts and stock price changes one month later.
Judging from the slope of the trend line, while operating earnings growth has a marginally greater ability to forecast stock prices than GAAP earnings growth, that outcome very likely stems from the enormous hype our three full-time financial networks spend on earnings season and the hit-or-miss-by-a-penny game. The results seem to indicate what one might expect: there seems to be volatility in the relationship as companies try to manage expectations and then have to take some of that back because they have drifted too far from reality.
Regarding free cash flow growth, the more sharply rising trend line shows a relatively more positive response of stock prices progressively as free cash flow grows. Based on our experience, we believe even more interesting ways than this simple approach can be found to build on the use of free cash flow.
Sometimes the results are startling. In the 2012-2013 election cycle (mid-2012 to mid-2013) there were some very interesting results just in the small sample of the 30 stocks in the Dow Jones Average. The Dow was up about 16%. Microsoft, which reported a significant jump in operating earnings from June 2012 to June 2013, saw its stock rise less than the Dow. On the other hand, Proctor and Gamble’s operating earnings dropped almost 50%, but its stock appreciated 27%. Microsoft’s free cash flow was actually flat for those 12 months, while P&G’s free cash flow went up 18% (again, while its reported earnings declined.)
One of our favorite books on investing was published 60 years ago: “The Battle for Investment Survival” by Gerald Loeb. We are sure he could not have foreseen that he underestimated how true that characterization would be today. While the changes over the years have been great, we find that a simple upgrade in the way we measure management – using free cash flow – has not only stood the test of time, but provides a robust advancement for investors to more reliably engage in future battles.
Ted Theodore, CFA
Chief Investment Officer
Past Performance does not guarantee future results
The S&P 500 Index is an American stock market index based on the market capitalizations of 500 large companies having common stock listed on the NYSE or NASDAQ.
The Dow Jones Industrial Average (DJIA) is a price-weighted average of 30 significant stocks traded on the New York Stock Exchange (NYSE) and the NASDAQ.
You cannot invest directly in an index.
R2 is the proportion of variability in a data set that is accounted for by a statistical model.