In our alpha research, researchers are attempting to find any strategy that has statistically significant predictive power on asset returns. These strategies, or what we call “signals,” are designed by using such tools as mean-reversion, lead-lag effect, momentum, analyst rating information, news sentiment, etc. Fundamental analysis, regarded as the cornerstone of investment, is obviously a very important direction of alpha signal design.

When talking about stocks, fundamental analysis is defined as the technique that attempts to determine a security’s intrinsic value by analyzing underlying factors that may affect a company’s actual business and its future prospects. We are trying, via fundamental analysis, to answer such questions as: Is the company’s revenue steadily growing? How is the company’s debt solvency capacity? Does the company have good profitability as well as high earning quality? Does the company have enough liquid assets compared to liabilities, etc.?

On a broader scale, fundamental analysis refers to the analysis of the economic well-being of a financial entity as opposed to its price movements exclusively. One can perform the analysis on sectors/industries or even the economy as a whole instead of single stocks. Contrary to fundamental analysis, technical analysis – which is another major form of security analysis and an important direction in alpha research – focuses solely on the price and volume movements of securities without caring much about the fundamentals of the underlying factors.

The various fundamental factors could be either qualitative or quantitative. In alpha research, we are primarily looking at quantitative factors, which could be measured or expressed in numerical terms. As financial statements are the standard medium by which a company discloses information regarding its financial performance, we often use the quantitative information extracted from financial statements to design alpha signals. Much empirical accounting research has also attempted to discover value-relevant accounting attributes from financial statements in order to enhance fundamental analysis.

Main sources of financial statements are the balance sheet, income statement, and the statement of cash flows. The balance sheet, as a static statement, shows a “snapshot” of an organization at a particular point in time while items from the income statement and the cash flow statement measure the flows over a period of time. It provides a list of a firm’s assets and liabilities, as well as the difference between them, shareholders’ equity, which is the net worth of the firm. Debt level as well as relevant financial ratios such as debt-to-equity ratio, quick ratio, and current ratio could be derived from the balance sheet. This helps an investor understand the company’s debt interest payment, credit rating, and performance with industry average. Other red flags such as large decrease in reserve accounts, increase in inventory, or big increase in accounts receivable also could be exposed in the balance sheet. The income statement provides a measure of profitability over a period of time, where all earnings/profits and expenses/costs data could be looked at. Earnings before interest measure the operation profitability, subtracting any interest burden attributable to debt financing. Due to conservatism, expense items might also include some costs that are not directly related to goods sold during the current period, while some research and development expenditure might not be shown. Items from the statement of cash flow reports important components on the well-being of a firm since it shows the cash change during the year. If the statement shows the firm is not able to meet its dividends’ demand but is keeping the productivity of its capital stock out of cash flow from operations, or the amount of cash flow from operations is lower than that from investing, we could see a debt problem – which might be a serious warning.

In-depth analysis of financial statements gives us insight into a company’s current and future performance. For example, you may discover, if you study the accrual of earnings in several respects, it has perhaps been manipulated and gives a biased illustration of a company’s status. High earnings today could always be viewed as an indication for investors to expect high earnings tomorrow. And the earnings of a company is made up of two parts: the actual cash flow, which is generated from the company’s operation, and the accruals, which is calculated and decided by accountants, and thus leaves room for manipulation. To analyze the quality of accrual, it is worthwhile to look at a case study conducted by Sloan et al. (2011).

1. First, Sloan scaled earnings, accruals, and cash flow by total assets to compare firms of different sizes.
2. Second, Sloan analyzed the relationship between earnings and quality of accruals. The data used was accruals, earnings, and cash flows from sample firms, and ranked on earnings. Sloan assigned firm-years into deciles based on the rank of earnings, and calculated the average value of earnings in each decile. Then he tracked the results of earnings for the previous and following five years around the calculated year.
3. Third, he compared the results with a ranked one: ranking the value on accrual part of earnings.

The result shows that, when using the earnings to make a prediction, a firm is expected to continue to have high earnings performance for several years into the future if it has high earnings performance this year. But after ranking the accrual component, the power of earnings’ predictability performs worse. So we could see that the component of cash flow is much more powerful than the accruals. In other words, when analyzing the future level of earnings, it is more reliable to rely on earnings generated from cash flows, rather than on the accruals.

In order to gain some understanding of a company’s value and financial performance, we analyze the valuation ratios, which are mathematical calculations using figures mainly from the financial statements. Some of the most well-known valuation ratios are price-to-earnings and price-to-book. Some empirical research indicates that factors related to contemporaneous changes in many financial statement ratios can yield significant abnormal returns.

Besides the generic items from financial statements, any correlative information outside the format is submitted in the form of a footnote. Footnotes are informative and disclose critically important content to help investors get a better view of the company and make informed decisions. Detail on matters such as accounting policy, changes in accounting methods, long-term purchase commitments, pending or imminent litigation, and executives’ stock options can be found. Barely does any company expose its mistakes or difficulties in headlines or tables; hence reading between the lines of these disclosures gives the diligent investors an advantage. In some cases, financial disclosures are used by companies to hide the fact and the effect of changing accounting rules, which might hurt stock prices. Empirically, when there are quite a lot of new footnotes in financial reports, some red flags might be buried in the long paragraphs. Also if you get no revelations from reading these footnotes, chances are that the company is being intentionally obscure. Having the ability to detect early warning signs in the footnotes of the financial reports sets apart the elite investors from the average ones.

We also see quarterly conference calls as a tool for corporate disclosure. While the financial statements give a snapshot of the company’s past performance, the conference call gives investors an idea of the current situation and the expectation of future performance from management teams simultaneously. Some commentators would say that from the tone and the manner of the CEO and CFO, especially in the Q&A part of the conference call when explaining the significant deviations in performance with previous estimates, one can gather critical information for long-term models as well as technical indicators. Empirical evidence shows that the sentiment of the conference call can reflect earnings surprises in the following 60 trading days. Also some particular macro-economic variables could be used as powerful stock price indicators in the relevant industry, such as the correlated relationship between the oil price and the stock price in the transportation industry.

Analyst reports produced by the investment bank or stock brokerage, the “sell side,” are other informative sources. Since these reports integrate, analyze, predict, and build valuation models via past, fundamental information on a specific stock and the relevant industry, their projections, comments, ratings, and recommendations produce meaningful components. And since large institutional investors who can move the markets because of the volume of assets they manage, refer to analysts’ views, these messages are those that you do not want to pass up.

Compared with other alpha signals like price–volume-based signals, due to the low update frequency nature of fundamental data, fundamental alpha signals have lower trading turnover and lower stock coverage. On the other hand, fundamental information tends to be reflected in stock prices over a relatively longer period of time. The cumulative returns to the fundamental signals usually concentrate around the subsequent earnings announcement, and level off one year after the signal’s disclosure, indicating a large percentage of abnormal returns can be attributed to previous years’ earnings changes of the companies.

Fundamental analysis can also give us ideas on alternative stock classifications. For example, stocks can be labeled as either “value” or “growth,” based on the company’s financial performance. “Value” stocks refer to those at relatively low prices, as indicated by low price-to-earnings, price-to-book, and price-to-sales ratios, and high dividend yields, while “growth” stocks refer to just the opposite – high price-to-earnings, price-to-book, and price-to-sales ratios, and low dividend yields. Similarly, one can also generate classifications by using other fundamental factors differentiating one type of stock from another. These kinds of classifications, which are related to fundamental information about companies, allow one to better observe market behavior of different groups of stocks, and thus design better alpha signals.