## Friday, November 11, 2016

### Direct participation in the Indian equity market: First estimates of some basic facts

by Anurag Dutt, Renuka Sane and Susan Thomas.

In the emerging field of household finance in India, the establishment of basic facts is th early priority. We have always been curious about the direct participation of individual investors in financial markets. One of the key goals of the Indian financial markets reforms was to facilitate financial inclusion, so that a larger mass of the Indian public could benefit from participation in the formal financial system. This would allow ordinary citizens to get better returns on their savings than fixed deposits in banks, and be better able to manage the risks of shortfalls in available savings to tide them over bad times in the economy.

Traditionally, our knowledge about individual participation was limited to aggregative statistics released by the Reserve Bank of India. While recent data from the RBI shows that the Indian household participation has slightly improved, it still appears low compared to the overall population despite the last two decades of market reforms. At the same time, this data has many limitations.

In this article, we take a first look at data from ownership of shares as seen at the National Securities Depository Ltd. (NSDL), one of the two depositories of the country.

### Evidence about household finance from securities depositories

India has a system of multiple, competing depositories which allows investors to choose a depository where they would like to hold their securities. The two organisations are the Central Depositories Services Limited (CDSL) and the National Securities Depository Limited (NSDL).  All securities that are traded on exchanges are settled in electronic dematerialised form at depositories. All primary market activity now creates dematerialised holdings only, at the depositories. The bulk of household participation in the stock market works through these depositories.

NSDL is the larger of the two depositories. Roughly 90 percent of all securities issued are held at NSDL. The data that we study is anonymised information about holdings of securities in the NSDL accounts. There is no information about who the account holder is, or any feature about the account holder like name or age or gender. What we are able to observe is a limited set of fields such as the date of account opening, the daily holding in each account till December 2015, and the district and state of the account residence. In addition, we have information about the class of investor: institution (such as banks, corporations, mutual funds, FIIs, FPIs, etc.) or individuals. Using this, we ask the following questions:

1. How many individual investors participate in equity markets?
2. How large is their equity portfolio?
3. How well do individual investors form portfolios?

Many households have indirect equity exposure, through mutual funds, the NPS, investment products sold by insurance companies, etc. In this article, we only analyse direct ownership of shares.

### How many individuals do direct participation in the public stock market?

Table 1 presents the total number of accounts in NSDL as of 31 December 2015. There are 24.6 million registered accounts. Many of these accounts are empty. We define an active' account as one which has at least one holding that is traded on the NSE or the BSE. By this definition, there are 13.6 million accounts.

Table 1: Participation statistics

Number of accounts (in million)

Accounts

Registered accounts 24.6
Active accounts 13.6

PAN IDs

Registered PAN IDs 14.9
Active PAN IDs 10.4
Retail Active PAN IDs 4.8

An investor can hold multiple accounts, particularly if the investor is an institutional investor. In order to understand participation by distinct investors, we merge all accounts with the same income tax Permanent Account Number (PAN) number. This includes retail and institutional investors, and will help us estimate how many accounts are held per investor at NSDL. We find that there is a total of 14.9 million individual accounts, which is around 60 percent of the total of 24.6 million.

When we further focus on only "active" accounts, we get a total of 10.4 million investor accounts, which is around 42 percent of the total. Finally, when we segregate the accounts into institutional and retail categories, we find that find that there are 4.8 million `active individual'' accounts at NSDL.

If there are five persons in a family, then 4.8 million accounts may map to households adding up to 24 million individuals or 1.8% of India's population.

Our evidence is biased in favour of understating the true state of household exposure to equities as (1) We do not see holdings by households at CDSL; (2) We do not see holdings of the household through mutual funds, insurance companies and the NPS; (3) We are ignoring holdings by households of unlisted companies and (4) We do not see equity ownership by households outside India. Hence, 1.8% of the population is a conservative lower bound for the fraction of India's households who have equity exposure.

### How large are individual holdings?

One measure of participation is their extent of participation, which is measured by the value of investors portfolio.

 Figure 1: Density of portfolio balance of individuals

Figure 1 shows the density of the value of the portfolio in all accounts at NSDL.  About 50 percent of these accounts have values of less than Rs.40,000, while 75 percent of accounts have portfolio value less than Rs.250,000.

The average value of holdings is Rs.1.33 million. Distributions of wealth always have strong skew: there are a few households with a lot of equity wealth and a large number with very low values.

There is an odd spike in the distribution. It pertains to portfolios of value between Rs.911 and Rs.1113. There are 256,297 accounts in this range. The mean number of securities in these accounts is 1.09. The story seems to be that 210,732 accounts have a few shares of Reliance Power, going back to the IPO on 15 January 2008. The accounts seem to have been dormant from 31 March 2008 till today. The rupee value of all these portfolios, put together, is negligible.

Our evidence is biased in favour of understating the size of the equity ownership by households as: (1) We do not see holdings by the same households at CDSL; (2) We do not see holdings of the household through mutual funds, insurance companies and the NPS; (3) We are ignoring holdings by households of unlisted companies and (4) We do not see equity ownership by households outside India. Hence, Rs.1.33 million is a conservative lower bound for the equity exposure of the average person.

### How wise are these portfolios?

A question that is often asked about direct participation in equities markets is the quality of the thinking. Equity investment yields good returns when a diversified equity portfolio is held over the long term. Wisdom in personal financial management lies in having high diversification and long time horizons. In our one snapshot of the portfolio holdings at NSDL, we look at the degree of diversification, by counting the number of securities in the portfolio. Every security is identified using an International Security Identification Number (ISIN).

 Figure 2: How many securities are owned by each individual?

Figure 2 shows that individual investors seem to have very few securities in the account. Around 29% of accounts hold just one known public equity ISIN, and more than 50% of accounts hold less than three known public equity ISINs. This suggests a high degree of underdiversification.

We further estimate the under-diversification value of the accounts using a regression approach of investor portfolio returns against the market index as the benchmark, well-diversified portfolio. This approach gives us an estimate for under-diversification as how far the regression R-squared is from one. This value can go from 0 to 100, and the lower the under-diversification value, the better the quality of the portfolio.

Another measure of portfolio diversification that we obtain from this approach is the value of the the portfolio beta. Generally, a well diversified portfolio should have a beta that's closer to 1.

We find that the average under-diversification is 93.760% and average portfolio beta is about 0.57. These indicate very low levels of diversification and poor quality of equity investment, and is consistent with most accounts being under-diversified because they hold very few equity securities.

Our evidence is biased in favour of understating diversification as: (1) The household portfolio may be made up of multiple individual accounts, while we have only aggregated accounts by the same PAN; (2) We do not see holdings by the same households at CDSL; (3) We do not see holdings of the household through mutual funds, insurance companies and the NPS; (4) We are ignoring holdings by households of unlisted companies and (5) We do not see equity ownership by households outside India.

#### Conclusion

We have uncovered three interesting and new facts. Atleast 1.8% of India's households are direct participants in the equity market. Their average equity portfolio holding is above Rs.1.33 million, and it is significantly under-diversified.

Anurag Dutt is a data scientist at the Finance Research Group at IGIDR. Renuka Sane is a researcher at the Indian Statistical Institute, Delhi. Susan Thomas is a researcher at the Finance Research Group at IGIDR.

1. The conclusion of the article that present equity investors are "not wise" appears to be questionable. The methodology appears to be in-appropriate to reach such an over-sweeping conclusion.

The conclusion is based upon the premise/assumption that a wise investor holds multiple securities from diversified sector. Hence, for an wise investor, the portfolio return should match or exceed the return of an hypothetical portfolo which is well diversified.

The above criteria has been tested by a frequentist regression methodology (i.e. by regressing portfolio return of 4.8 million individual returns on the return of the hypothetical portfolio).

The results of the frequentist regression methodology is obviously going to be impacted by 50% of the sample which has a portfolio value of less than 40K INR. It appears very unrealistic to assume that with such low value portfolio, an investor will have multiple equity investment.

The results would be more robust if analysis is limited to high portfolio value individuals (say more than 250 K INR).

Otherwise, the over-sweeping conclusion appears to be simply a reflection of the extremely skewed portfolio value distribution rather than to do anything with investor intellect.

2. Nice analysis. Just one point: Average equity portfolio value is Rs.1.33 million. What is the median value?

Please note: Comments are moderated. Only civilised conversation is permitted on this blog. Criticism is perfectly okay; uncivilised language is not. We delete any comment which is spam, has personal attacks against anyone, or uses foul language. We delete any comment which does not contribute to the intellectual discussion about the blog article in question.

LaTeX mathematics works. This means that if you want to say $10 you have to say \$10.