Sorry, this number is not declared. But I was wondering whether the size of the Margin lending book could help serve as a proxy? With the new SEBI margin lending rules, all brokers are forced to provide same maximum leverage. This must necessarily imply some lower bound on the assets that client is bringing for their trading accounts. The big leap of faith here is that trading account assets serve as a proxy for all assets, on an average, across brokers. This could very well be false. With this assumption, one can compare the margin lending book across brokers.
This data is thankfully declared by NSE on a monthly basis. The margin funding book is volatile and one should ideally take an average across multiple months.
margin funding.xlsx (27.9 KB)
One would actually be surprised to learn that Margin funding per client is higher for Angel, than even Kotak and ICICI bank. On the overall basis, Kotak’s book is 2416 cr, ICICI bank’s is 2014 cr and Angel’s is 1759cr.
Do you have any data to back this claim?
Throughput per client, i agree (because by their very definition, HNI are high net worth). Throughout on an overall basis, this logic would not apply. If Kotak has 1000 HNI who they can cross sell to with say, 1 cr each. And Angel has 1 lakh retail investors with 50k/year each that they want to invest in Mutual funds and 10k worth of yearly premiums that they want to spend for their insurance, we quickly start to see that the volume game can work too. I personally do not find such macro arguments (without data) very useful in my investing because without data they are simply opinions. One would need to look at the precise distribution of number of clients versus net worth for all brokers in order to make any data backed statement. For some value of average affluence, dominating volumes would enable higher cross sell than focussing on the head of the distribution (HNIs). Without the actual data I do not know which side of the argument is factually correct.
Per client basis, completely agreed. But Zerodha and Angel are playing the numbers game. Angel’s stated aim is to become the largest brokerage in India. Whether they can get there, is a different question altogether. One would need to know the distribution of income versus how many people have that income for each of the brokers in order to comment on how much HNIs contribute to the cross sell. Also, as the margin trading data already shows, it might just be a preconceived notion that HNIs invest with banks and people without much of a net worth invest with discount brokers.
There is another problem with using total assets as a proxy for cross sell. If the cross sell has not been much of a focus area for the company yet then the average total must also be necessarily low (because the cross sell specially for MF) has not happened. This is indeed the case for Angel. ICICI and zerodha have been focussing on cross selling. Angel, not so much. Thus, comparing on total assets would present an incorrect picture because one of the competitors in the metaphorical race is running with 1 leg tied.
As an aside to the angel investment discussion, i personally don’t find much merit in the argument that bank backed brokers are safer. Balance sheet is equally safe with all brokers. There might be a perception of risk, but could just be a preconceived notion as far as my understanding goes. The underlying investments are with the depository. The broker is just a UI/UX. I don’t know what the limit for HNI is (whether i am one), but I have 60-70% of my assets with zerodha and intend to keep it that way throughout my life. 
UX over everything else, when selecting a broker.
Disc: Invested, biased.