The chart below depicts the global Google Search Trend for the word "Factor Investing" over the period January 2004 to January 2022.
As visible, the interest towards factor investing over time has increased exponentially over the last couple of decades.
This chart depicts the growth in the NAVs of S&P’s factor based indices vis-a-vis that of the S&P 500 Index over the period 4th July 1995 to 31st August 2022. All the NAVs are in USD and have not been converted to INR. All the indices have been scaled to 1,000 as of 4th July 1995.
Past performance may or may not be sustained in future and is not an indication of future return.
According to a survey conducted by FTSE Russell in 2019 to evaluate the prevalence of smart beta/factor based strategies among 178 asset managers spanning across various AUM tiers and regions, over 58% of the respondents have adopted smart beta strategies in their portfolios. In addition, of those who were still evaluating such strategies, more than half indicated plans of incorporating them in their portfolios within the next 18 months. Interestingly, the growth of multi-factor smart beta strategies has been especially dominant, with 71% of managers using multi-factor strategies in 2019, as opposed to 49% in 2018. It is also worth noting that over 60% of the surveyed respondents have adopted smart beta strategies to make long-term strategic allocations as opposed to a mere 7% asset managers that primarily use smart beta for short-term tactical purposes. (FTSE Russell, 2019)
Factor investing, as an investment discipline, is a unique amalgamation of academic theories, empirical evidence, and data analytics, making it vulnerable to the quality and quantum of data available. Conventionally, investment professionals and academicians have evaluated and identified factors primarily using market (price and volume) and fundamental (reported financial statements) data. Traditional data for investment analysis, both market and fundamental data, is produced in a structured form, thereby enabling financial analysts and professionals to run linear statistical models, such as regression analysis, to gauge the efficacy of different investment factors, particularly the style factors. Although a straightforward approach to making investment decisions, relying solely on market and financial data has its pitfalls, largely because it fails to analyse the data stored in unstructured formats, such as social media data, textual, audio-visual data among others (Melas, 2022).
With an exponential and constant rise in the 4 V’s of Big Data, namely Volume, Velocity, Variety, and Veracity, it is necessary to effectively analyse the data stored in both structured and unstructured forms to make more informed investment decisions. Advances in information technology, particularly the increase in the computational power and cloud computing, is catalysing the adoption of advanced machine learning and artificial intelligence models. Techniques such as natural language processing are gaining ground in the efficient extraction and analysis of unstructured datasets.
Adoption of advanced analytics is now widespread in the domain of factor investing globally to improve factor identification and enhance portfolio construction and optimisation. By using advanced linear as well as non-linear models, investment professionals can cover very large datasets, enabling them to identify the ‘true’ nature of the various factors. These are being used to address the significant underperformance of the value factor globally over the past 10-15 years, which can be partly attributed to the low interest rate environment and differences in the manner intangible assets are treated by the new-age technology firms and conventional value firms. For example, research and development (R&D) expenses are generally treated as revenue expenditure instead of being capitalised as intangible assets, decreasing the book value of such technology firms (Melas, 2022).
Continuous research, using robust machine learning techniques, to address the deficiencies of traditional factor and investment analysis, will ultimately culminate into more diversified and optimised portfolios with superior factor exposures (Melas, 2022).
One of the main reasons for the lack of adoption of such strategies in India has been the relative lack of their availability. Also, the availability of data, across time as well as companies, has been very constrained. Over the last few years, NJ Asset Management has put together high quality data spanning more than 20 years to create a repository of daily factor scores for over 1,500 companies. This database, combined with our in-house data analytics capabilities have provided the foundation for our factor based strategies across our portfolio management and mutual fund offerings.
In terms of future possibilities however, we are just at the beginning of our journey. As computing power, data analytics and evidence based intelligence transform other industries, we are working on adaptive rule based strategies that we expect to define the future.
One of the key areas where we expect progress is the development of protocols that assign weights to individual factors based on prevailing market conditions. This will be a stepping stone to truly adaptive protocols that rely on data driven machine learning technologies to deliver a superior investment experience that is within the reach of all investors.
Consequently, we see investor acceptance of these strategies only increasing from this point with the availability of a diverse range of strategies. NJ Asset Management is the first exclusively dedicated rule based active manager in India and plans to lead this effort.
With almost all mutual fund investments held in discretionary active funds, factor based investments are a natural diversification opportunity. This eliminates human bias at the investment decision making stage along with time-bound rebalancing providing a markedly different approach than the currently dominant strategies.
Moreover, the share of the mutual fund industry in financial and total savings is still extremely low. As the economy matures and inflation experiences a structural reduction, the preponderance of fixed rate savings is also expected to moderate. With a share of more than 50% of financial savings (The Reserve Bank of India, 2020), this pool of fixed rate savings is expected to drive the growth of professional asset management in the coming decades.
NJ Asset Management believes that the success of factor-based strategies in generating positive excess returns in a cost-effective manner and the relentless research to make factor performance more consistent will make the adoption of these inevitable in India over the coming decades.