Contributed By: Rajeeb Bharali , CFA and Rajnish Kumar, PhD
Benjamin Graham is considered as the “Father” of Value Investing and his book “The Intelligent Investor” is lauded as the most influential book ever written on value investing. A value investing strategy implies buying stocks whose intrinsic value is greater than its market price. In one of the most pioneering works, Basu (1977) empirically tested the concept of value investing by examining price-earnings (P/E) ratios as leading indicator of future investment performance. This was followed up with the seminal paper by Fama-French (1992) where they used book-to-price to represent the value factor and that laid out the path for a quantitative way to capture the value premium as has been exemplified in the following years, perhaps decades, in the asset management industry.
To a keen observer, this begs the reader to ask 3 questions: 1. Is the Price to Book the most effective way to capture the value premia? 2. Is it wise to apply a standard set of valuation multiples uniformly across sectors/industries? 3. Why did the world of finance ditch Basu’s Price to Earnings in favor of Fama-French’s Price to Book ratio as the apt measure of Value? This article is the first of a series of 3 short articles which will try and address the above questions in the sequence stated.
The Value Premia
Over the last 12 years, value factor has under-performed the broad market. Numerous reasons have been cited to explain this underperformance. A low rate environment and substantially long expansionary phase leading to outperformance of growths stocks over value has characterized the past decade. At the same time, questions have been raised on the adequacy of valuation parameters used in capturing the value premia, especially with regards to growing intangible assets. A comparative analysis of total returns of different style factors shows that value factor has consistently underperformed the rest.
With changing dimension of industry, asset quality and performance of value factor, it is imperative that we examine the effectiveness of other price ratios in exploring other dimension of value factor. Academia as well as practitioners use different price ratios as a representation of value such as price-to-book, earnings yield, cash flow yield, dividend yield, price-to-sales, EV-to-EBITDA and different variants of which is either historical, forecasted or average value of fundamental variables in the price ratios. As stated earlier, price to book are dividend yield are the favourites, especially amongst quants trying to exploit the value premia. But can these 2 variables capture the value premia in its entirety. To ascertain this we set out to assess each valuation multiple in isolation and see its efficacy over time.
As a starting point, we consider MSCI USA Index as our base universe and consider data from June 1994 till May 2019. We collate the valuation data for each of the securities forming part of the universe across the mentioned time period. As part of the outlier treatment, we winsorize the data on both ends (remove top and bottom 2% of the universe by each metric). Using each valuation multiple, we then divide the data set into quintiles. For example, based on Price to Earnings multiple, we rank the universe of stocks in ascending order and quintile it so that the top quintile (Q1) contains the stocks with the lowest PE multiple (value) and the bottom quintile (Q5) has the highest PE multiple (growth) stocks. In order to generate the pure price to earnings return stream, we calculate the return spread between Q1 and Q5. Similar exercise was done for all the other valuation metrics under consideration – Price to Book (PB), Price to Sales (PS), Price to Cash Flow (PCF), Price to Dividend (1/DY) and EV to EBITDA (EV_EBITDA)
The cumulative returns of these hypothetical long short portfolios (by multiples) are shown below. As can be seen, there has been a wide and growing disparity between PE and PB (or DY). In fact, for the period considered, DY has been negative and PB being close to neutral. PE has outperformed all other metrics hands down. Also to note is the fact that each valuation multiple has led to different outcomes, although correlated to a large degree.
While the above chart implies dominance of PE multiple, it is also worthwhile exploring its performance on a risk adjusted basis. The below table shows the performance analytics of different valuation measures (Sharpe Ratio is calculated assuming risk free rate of 0%):
Consistent with cumulative returns chart, Price to Earnings looks more attractive as compared to other valuation metrics. Price to Earnings also performs better on a risk adjusted basis as compared to MSCI USA Value.
While the academic literature gravitated towards a single measure of value, Price to Book, and was ably mimicked by practitioners (mostly quants), it is, by no means, a complete measure of value factor premium. Although, many still use a composite value scoring methodology, encompassing most of these valuation parameters, it raises questions on a blanket approach being adopted across sectors/industries. For example, Price to Book may be an excellent measure for Financials; it is of limited use in the Technology sector. Efficacy of either PB or DY also raises the fundamental question of whether it takes into account the earnings power of a firm. A stock may have a very attractive PB or DY multiple but if it has a low or even negative earnings (historical or forecasted), it is best to stay away from such stocks and avoid value traps.
In our next article, we will assess the sector specific valuation multiples. We will create a new composite value indicator which incorporates sector relevance and compare it with an equal weighted composite of all metrics applied uniformly across sectors. It is very important to accurately define factors which capture the essence of the premia in its true sense, in times of changing economic environment.
About the Authors-
About Rajeeb Bharali, CFA:
Rajeeb is a seasoned investment professional with ~9 years of experience in the field of investment management. His primary area of work is into asset allocation/portfolio construction for Fund of Funds (FoF) and investment manager research and selection. Rajeeb holds a BS in Engineering from Birla Institute of Technology, Mesra, graduating in 2008 and is an active member of the CFA Institute as well as the CFA Society India.
About Rajnish Kumar, PhD
Rajnish Kumar has over 8 years of experience in factor research across equity and fixed income. His area of interest includes ESG, smart beta, portfolio construction and risk-return attribution analysis across different asset classes. He has publications in top tier journals. He holds PhD in Economics from Indira Gandhi Institute of Development Research, Mumbai.