Quantitative Investment: Using The Language Of Numbers To Uncover Investment Opportunities by Balint Anton Francisc
Value Investment, as a philosophy, has moved a step forward when Warren Buffett adopted from Charles Munger’s approach towards quality. Mr. Munger’s investment philosophy was based on Philip A. Fisher’s believe of outstanding businesses that become of such nature as a result of both qualitative and quantitative features. Benjamin Graham and David Dodd in Security Analysis have accentuated vividly that qualitative considerations, such as the superiority of the relationship between management and its employees, the loyalty of the customers and the strength of the sales organization within a company ought to be left out when valuating an investment opportunity as they cannot be monetized.
For decades professionals from a wide range of industries and fields of knowledge have been trying to pin point what is the mental process through which great value investors go in order to uncover investment opportunities, decide how much capital to allocate in them, how to make accurate decisions and when to admit defeat. In other words, the world around these great names of investment is asking: What is the checklist that will ensure long-term success in the game of capital allocation? In the last five years, we have received an answer coming in the form of a book - Quantitative Value by Wesley R. Gray and Tobias Carlisle and reinforced by the recent Deep Value by Tobias Carlisle.
The ideas upon which the books are constructed come from the need and necessity to identify a mathematical formula that imitates the investment process of these legendary names. Joel Greenblatt, with his Magic Formula, set out to do just that and it outperformed the market outstandingly well since he started applying it. However, as we shall see, certain metrics outperform even the Magic Formula.
Charlie Munger: Invert And Use “Disconfirming Evidence”
This article will examine the metrics underlying the strategy called Quantitative Investment. We will look briefly at what the Magic Formula is and what are the issues with it and then we will move towards assessing some of the most important research that resulted in what Tobias Carlisle calls the Acquirer’s Multiple.
The Magic Formula: Composition and Faults
Greenblatt’s Magic Formula ought to have been the mathematical expression of Warren Buffet’s definition for a wonderful business or a franchise. The arithmetic rationale roots itself in Buffett’s suggestion to use return on equity capital as a measure for a good business, as stated in the ‘Shareholder Letter’ in 1977. Therefore, the algorithm reads as following:
Return on Capital (ROC) = Earnings before Interest and Taxes (EBIT)/Capital
Capital is defined for simplicity purposes as fixed assets + current assets – current liabilities – cash (Joel Greenblatt, The Little Book that Beats the Market, 2005). Consequently, the formula ought to measure the efficiency of capital employed and allocated by the management: the higher the ROC, the better the business because the more money ought to have been earned per dollar of capital employed.
However, there is a second component for the Magic Formula: the earnings yield, identified as EBIT/Total Enterprise Value. The TEV is defined as market capitalization + total debt + preferred stock + minority interests – excess cash. The idea behind the second component is to identify a bargain. Therefore, when using the two algorithms to rank stocks, first by their ROC and then by their Earnings Yield, we should get a market-beating strategy. In fact, according to the study conducted in Quantitative Value, Chapter 2, the portfolio selected based on the Magic Formula, for the period between 1964 to 2011 outperformed both the S&P 500 and the returns from the 10-year Treasury bond: the CAGR (compound annual growth rate) of the stock universe selected based on Greenblatt’s mathematics was 12.79% for the period, while the S&P 500 returned 9.52% CAGR and the 10-year Treasury returned just 7.52% CAGR.
In quantitative investment, the metric you use is crucial for getting the right data and avoiding useless information that can drag down the portfolio performance. Astonishingly, in both Deep Value and Quantitative Value, the authors found that the ROC component of the Magic Formula actually drags its performance down. The fault lies in the mean reversion nature of the qualitative measure, ROC, while the price ration (EBIT/TEV) remained a more stable indicator of stock performance. The mean reverting feature of the ROC factor means that investors tend to overpay for quality should they select stocks primarily based on the Magic Formula.
Fortunately, Greenblatt’s formula can be improved:
The Acquirer’s Multiple: The numbers, the competition and the mean reversion
“Many shall be restored that now are fallen and many shall fall that now are in honour” reads the famous quote from Horace in Ben Graham’s Security Analysis. These words are a powerful example of mean reversion and of what the market really is: in short-term it is a voting machine but in the long run, it is a weighing machine. This means that value, in any shape or form, regardless of the prices paid for the asset in question, sooner or later will align with that price.
In Quantitative Value, the authors analyzed six price ratios in the hope to find the best indicator of finding an undervalued company: remember, quantitative investment is the practice of trying to apply well established value investment principles (such as economic moats and the margin of safety) by analysing data. Therefore, indicators that allow reading the data are crucial for success. This is what pricing ratios are: signals that a company might be undervalued by the market. The authors studied the following:
Earnings Yield: E/M (E=earnings before extraordinary items – prefer dividends + income statement deferred taxes and M=market capitalization) –this is the inverted P/E ratio.
Enterprise Yield: EBIT or EBITDA/TEV (TEV = Total Enterprise Value) – the idea behind using EBIT or EBITDA is to get a sense of earnings that are not affected by temporary losses or gains from extraordinary events and the TEV allows the investor to take consider the total cost of the acquisition not just the cost of equity.
Free Cash Flow Yield: Free Cash Flow/TEV (Free Cash Flow= Net income + Depreciation and Amortization – Working Capital Changes – Capital Expenditures) – this ratio allows to manage non-cash transactions and reflects the efficiency of decisions made regarding capital expenditure.
Gross Profit Yield: Gross Profit/TEV – the Gross Profit it is said to be the clearest picture of income because the lower you dive into the Income Statement the more tainted the income figure becomes.
Book to Market: B/M (B=Common Equity + Preferred Stock par value or assets –Liabilities-Preferred Stock redemption value + Balance Sheet deferred taxes) – here the argument was that B/M ought to be a stable metric for identifying a solid business by gazing at its balance sheet valuation.
Forward Price Estimate=FE/M (FE=the consensus of future earnings within the international investment community, usually found in the analysis of institutional investors) – obviously, it is better to use the future than the past to predict the future.
By testing these pricing ratios, the authors used an immense stock universe: all stocks listed on the NYSE, AMEX and Nasdaq and measure their returns from 1964 to 2011. They found that the best performer that beat the S&P 500 was the EBIT/TEV, followed by EBITDA/TEV. These two ratios also performed the best when they were scrutinized under the Absolute Measures of Risk, the Sharpe and Sortino ratios: they both are risk/reward metrics. However, the Sharpe ratio examines the historical relationship between excess return (return in excess of the risk-free rate) and volatility (risk); while the Sortino ratio measures the excess return over a minimum acceptable return per unit of downside risk (the extent to which the portfolio has fallen in the past).
The cheapest socks based on the winning ratio tend to outperform the more glamorous, or expensive stocks identified with it – the reasons root themselves in the fact that companies with their prices painted red are thought as walking dead and therefore, the competition exits the market hand in hand with the investors. However, as soon as the competition diminishes, these companies that are ‘poor earning winners’ revert back to the mean (the average stock price) because they have market space (monopoly) that boosts their economic power (they can increase prices without losing customers).
Now, the idea here is not necessarily to make blind acquisitions of stocks based on the cheapest company resulted from EBIT/TEV. The rationale is to use this as a primary screening tool. The EBIT/TEV ratio has to be used in concert with a whole range of other signals, some of which are qualitative in nature. For example, uncovering a relatively cheap enterprise based on the Acquires Multiple (EBIT/TEV) that declares a buy-back in the near future should not lead to automatic bias towards buying the stock – the nature of the buy-back has to be qualitatively analyzed: is the company truly undervalued? How much of the stocks will be purchased by the company? Does the management have a solid record of keeping its words when announcing a buy-back or is this a strategy to temporarily boost the stock price? All these questions require the Philip Fisher ‘detective approach’.
Finally, be aware of data errors. For example, investors using data tend to include only listing stocks (survivors) and therefore, they get an erroneous picture of their performance. Try to include non-listed companies as well. It will be helpful to use the Center for Research and Security Prices database to help avoid this bias. Also,