(Update, October 10, 2019, 8:35 p.m.: This article has been updated with more specific numbers on IBKR's U.S./non-U.S. client split. The original version used an estimate based on the company's previous guidance.)
IBKR has been indiscriminately sold in the zero commissions price war it itself sparked. In my opinion, the market reaction is overly US-centric for this global broker. One of the world's most profitable and consistent growing franchises is on sale at a P/E consensus 2019 ratio of 21 (this number reduces to 14 excluding IBKR's excess capital).
We will analyze the immediate and then long-term implications of the US price war for IBKR. First, a quick recap of the price war timeline is in order.Price war timelineBeginning of August: Fidelity starts offering higher rates on client cash balances with an automatic sweep to money market fundsIBKR starts its formula “IBKR Lite” in the last week of September. IBKR Lite offers free tradingonly for US individuals trading US stocksIBKR will – for the first time ever – sell order flow to external market makers (e.g. Citadel, VIRT) for clients that switch to IBKR Lite. Other major brokers have been earning both commissions (8-10$ per trade) and from payment-from-order-flow “PFOF” for years.Schwab matches this offer by announcing zero US stock commissionsAmeritrade and E-Trade followImmediate earnings impact
How will IBKR earnings change if a client changes from IBKR Pro to IBKR Lite?
Recent media reports unfairly categorize IBKR in the same camp as AMTD and ETFC as far as earnings sensitivity to the price war is concerned. This is wrong on three levels:
Relevant client base exposed is much smaller than peers
The market seems overly US-centric in judging the damage for IBKR. This is however more relevant to judge future implications of the US price war (see next section).
Relevant affected revenue is five times smaller than widely suggested
In recent media, something along the following lines was reported:AMTD/ETFC/IBKR each earn ~33% of revenue from commissionsSCHW and Fidelity “only” earn 8-10% of revenue from commissions
This argument neglects the fact that only 20% of IBKR commissions are indeed affected US stock commissions. I calculated this based on the monthly press release on execution quality, in which IBKR discloses commissions on Reg-NMS US stock trading. The remaining 80% of IBKR commissions are from foreign and other products trading. In contrast, peers are almost 100% US-based. Multiplying 20% and 33% of revenue yields only 7% of total revenue that is relevant affected revenue for IBKR. This includes US trades by foreigners which are not affected (this price war only benefits US individuals). Given the 31% US and 69% foreigner split, we can safely assume less than 3% of IBKR’s revenue is relevant in the discussion. Now, will that affected revenue actually shrink?
US Commissions loss is – as opposed to peers – compensated by PFOF
Based on AMTD, E-Trade and Schwab 10-K filings, both players earn both commissions and payment-for-order flow. Payment-for-order flow revenue is consistently and almost exactly three times smaller than commissions revenue. Simplistically, as we know that 2018 commission per trade for Ameritrade and E-Trade were approx. 8$, we can impute that these players earn on average 2,5 $ per trade from PFOF.
Using 2,5$ per trade as an indicator for IBKR Lite trade revenues would be misguided, however. Actual PFOF is negotiated on a cents per stock traded basis. The revenue per trade is therefore affected by the number of stocks traded and hence the trade size. An average account at Ameritrade and E-Trade is somewhat larger than 100 k$, while an IBKR client is about 250 k$ in size. However, I only expect unsophisticated clients and/or very small clients to switch to Lite, as this is advantageous at a very small trade size because of the absence of a fixed minimum commission cost.
Assuming a smaller trade size for IBKR Lite clients would suggest 1-2$ per trade in PFOF revenue per trade. Interestingly, IBKR Pro charges only 1$ for these same sized trades. We can hence safely assume that IBKR revenue will not be affected negatively from incremental clients switching to IBKR Lite. Recently, IBKR founder Peterffy confirmed this on Bloomberg TV, see here: “We do not expect a revenue change”.
Note that another incremental revenue driver for IBKR are the lower rates on cash and higher margin loan rates in the IBKR Lite offering.Long-term impact
Existing client base
Individuals & Advisors within American client base
Existing IBKR clients can switch to other US brokers which are now cheaper. Note that the offering is only for US stocks, and, importantly, US persons. The relevant client base is hence US individuals and US advisors. I add advisors because Schwab is adding free commissions for trades made by advisors on behalf of Schwab clients.
Only 49% and 16% of IBKR clients are respectively individuals and advisors. We have to make an educated guess about how many of those segments are actually US residents. To my great surprise, as of the Q3 2018 10-Q filing, IBKR started disclosing the exact percentage of non-US accounts and it is very high: 69%. In the past this disclosure read persistently more than 50% . Since institutional clients are probably biased towards the US, I would assume that only 20% of individuals are US residents. To make the weighted average work with the disclosed 69%, institutionals would have to be 40% US based. For the purpose of conservatively estimating the price war exposure however, I will assume that advisors are unusually biased towards the US, i.e. 60% of advisors are US based. Altogether, this means that only 19% of the IBKR client base is relevant in the discussion of the US price war (i.e. 49% x 20% + 16% x 60%). Keep in mind that the revenue impact of this group is higher as US clients are richer in general and international accounts are the result of more recent growth. Newer accounts tend to be smaller, and for IBKR in particular also less active traders.
Within this group, there will be clients at risk of churning, who wish to have a more comprehensive client service and do not benefit from IBKR’s unmatched differentiated capabilities in derivatives and foreign markets.
Because the brokerage industry is characterized by high switching costs and low churn on the one hand, and the fact IBKR has industry leading rock bottom churn on the other, it is reasonable to assume in a bear case that 10% of this relevant group will leave.
If 10% of the relevant group of US individuals and advisors leave, this would mean 1.9% of the total IBKR client base. At the historical client growth rate of 16%, this represents 1,4 months of client growth.
Future client growth
Individuals & Advisors within American client base
As discussed in the previous section, the relevant client group affected is only 19% of the total IBKR client base. Because client growth is faster abroad, the relevant segment of growth affected is even smaller, but we conservatively neglect this fact. If we assume that 50% of growth in this segment disappears, this would slow IBKR’s growth by 10%. With a prior 16% CAGR growth in clients, this would still mean client growth of 14.5% CAGR.
Note no positive effects were considered from the overall market growing when prices for a product go to zero (i.e. commissions), neither from the public peers group taking a higher share of overall market growth from Robinhood. These effects are hard to measure but by no means immaterial.
IBKR was first mover in this price war. This suggests IBKR founder should at least have thought about the scenario of unleashing a price war with some negative effects on growth in US individuals and advisors, as I calculated above.
So why did Peterffy do this?
The real victims in this commissions price war are small independent brokers. The largest of “small brokers” are E-Trade and Ameritrade. Many are predicting a merger between the two. Peterffy expressed on Bloomberg however that it does not necessarily make sense for E-Trade to merge. A merged entity would still be undifferentiated and uncompetitive versus Schwab and Fidelity.
What then does Mr. Peterffy suggest? I believe another option for many of these smaller players is becoming an introducing broker using the very cheap and ultra-scalable IBKR platform. This is where the opportunity lies for IBKR, a player with an ultra-scalable platform and a small niche client base that is however very active and profitable for the company.
Interestingly,beforeScottrade wasbought by Ameritrade in 2017, it had decided to become an IBKR introducing broker. This was in the context of slight cost pressures and commissions coming down from 10$ towards 8$. For context, Scottrade had 3,5 million clients at the time, five times more than IBKR’s total tiny base of 0,6 million clients.Market reaction fuelled by representativeness bias
To come back on the point of IBKR having penetrated only a very small share of total addressable market, I suspect part of the abrupt market reaction is based on the representativeness bias. The first representativeness bias is the US-centric reaction of a US holder base in the most global broker of the peer set. The second representativeness bias is more subtle.
Because US brokers have cut their commissions to zero, many market participants believe that other brokers are absolutely the better deal now compared to IBKR, because of the great client service at peers. How will IBKR stoke growth in the US now?
As an IBKR client myself, I will invert this argument: all other brokers are the same with client service handholding but with a severely limited investing offering (either very expensive or no foreign trading /derivatives/algorithmic trading capabilities). IBKR is the only broker for certain sophisticated investors and traders.
It is indeed easy to imagine that for most people, IBKR will not be the best option in the US. However, as an investor, the relevant question is how fast will IBKR grow in proportion to its existing tiny but very active client base. Indeed, the client bases of Schwab, Ameritrade and E-Trade are respectively 22, 17 and 7 times bigger. For US based clients, Schwab is probably 50 times bigger. My point is that it is easy to imagine “the median US brokerage client” as an investor in brokerage stocks – hence the stock market reaction – but the median US brokerage individual does not matter much to IBKR investors.
I believe the market is overly pessimistic because of representativeness bias. IBKR is growing worldwide from a tiny base of niche customers that do not want handholding and want extensive trading capabilities (i.e. the freedom to trade foreign stocks, different products, programmatically).ConclusionRelevant US Stock Commissions affected are immaterial for IBKR (<3% of revenue). Nevertheless, client switches to IBKR Lite will be revenue neutral, as confirmed by analysis and recent founder comments.The existing client base is incredibly sticky, many times more than the already sticky industry. The maximum exposure in terms of one-off client switches is about 1.9%.IBKR’s steady state client growth could slow from 16% to 14.5% CAGR in a bear caseIBKR is growing from a very small niche base of very active traders and investorsIBKR does not need to be the best broker for your typical US brokerage clientIBKR is a very scalable platform that could win large introducing broker contracts because of the recent price war pressure on smaller brokers who are the real victims (see the example of Scottrade which would have septuplet IBKR’s total client base). This positive optionality is not priced into the stock today.
Disclosure: I am/we are long IBKR. I wrote this article myself, and it expresses my own opinions. I am not receiving compensation for it (other than from Seeking Alpha). I have no business relationship with any company whose stock is mentioned in this article.