No transaction costs
August 18, 2009 in Finance
In stock market trading, a major inconvenient that bothers investors are transaction costs. What about creating an investing website which would charge no direct transaction fee, but charge a percentage fee on winners only?
- Investors would not need to face the constraint of realizing a performance that compensates transaction fees, which would attract many investors,
- They would probably be more willing to give up a percentage when they win, provided that they won! They would not loose anything when they loose.
Note: You get transaction fees from purchases and winning positions sells. Therefore, only less than 1/4 of revenues are cancelled (since markets tends to go up) while 1/4 of charges are increased.
joan said on August 18, 2009
Yes man, this idea is so cool. But you’ll have to charge twice as much in order to compensate your costs for all the loosing transactions… You can also hide your fees by selling at a price slightly different from the market price, like the forex brokers.
henry said on August 19, 2009
yes you’re right… But let’s avoid criminal ideas haha. I’m trying to find a way to avoid these fees which really are a big problem for investors. I was thinking that charging twice as much the winners could attract more people than charging the same for everyone. But formulated like that that idea sounds like comunism..
paul said on March 14, 2010
I’m not sure such a system could work for many reasons :
First of all : A broker is a party that mediates between a buyer and a seller ( Wikipedia ), so it should have a neutral position. If it’s remunation in based on the performance of some investors then it might be motivated to give them better price (on the bad investors back). More over the broker would be exposed to the market, since it’s overall revenue would depend on the investors portfolio.
To finish this system, which is a social one, means that good investors support the bad investors brokerage fees. If I was a good inverstor.., I would rather pay cheaper and fixed commissions then support the losers.
Joan said on March 14, 2010
Nice comment, couldn’t agree more. Your idea just got spanked henry!
Henry said on March 15, 2010
Thanks for your interesting comment Paul. Let’s discuss it a little bit.
First of all, the wikipedia definition. Why should a broker be neutral? I doubt the existence of a law article that prohibits such a system. Brokerage is a competitive market where everyone is free to pick up his own strategy and competitive advantages. Most people have a medium risk aversion profile, which means that (I’ll mention it for the people who don’t know what an aversion profile is) they are more willing to take low risk and get a medium revenue rather than a lot of risk and a high revenue or no risk and low revenue. If these investors are not certain to win money on their positions, they can totally be interested in an adaptative system, which strictly aims to minimize the standard deviation of profits and losses. If you are not willing to support the losers, maybe some others are. If they are certain to win … Wait, who is?
For your last argument, who is a “good investor”? This does not mean much nowadays. If fundamental analysis still made sense I would agree with your argument but the way markets work definitely changed: some people invest huge amounts of money, which allows them to control the price of stocks, some others have access to this information and can make a way out of it, but most investors (probably 90%) only have access to public information. In other words, among these 90%, I don’t believe there really are good or bad investors. And you can’t assume that more than half of your customers invest blindly in a stock, without collecting information. Such a system could definitely interest these people.
Furthermore, the broker could not give winners a better price since it goes directly against the initial purpose of its system which is, as you mentionned it before, social (or sort of because Stakanov wouldn’t have invested a cent on the stock market). I would call this system a risk moderator, inbetween the fixed income market and the traditional stock market. There is no such incentive. The fee is determined based on a combination of the amount of money invested and the performance and that’s all. Nothing’s on the bad investor’s back; he is just not charged.
And according to me the argument of the market exposure is true for everything and everyone. As investment slows down (like during a crisis) traditional brokers suffer of the market too. Just as the hot-dog vendor or the small entrepreneur who lost their jobs in march 2009. But I must admit one fact: during a crisis, although your customers love you, your revenue decreases more than for a traditional broker. Otherwise, except during a crisis, markets tend to go up, and so does the broker’s revenue. Now how can we make sure that we have more good investors than bad ones (that we have a good investors’ portfolio). Well I already gave my opinion on it: there are no good or bad investors. We’re left with the market risk exposure.
Please address.
Jonas Romain Wiesel said on March 15, 2010
A few numbers wouldn’t hurt that argumentation, I cannot say personally. What is the average brokerage fee on stock markets? What is the average return of these markets, can we predict it and with what interval of confidence? Then with those probabilities, costs and returns, what is the likely hood that this model is profitable for both parties?
Henry said on March 17, 2010
Joan, you would probably need to charge less than twice as much the winning positions to compensate the loosing ones since transaction costs are lower on smaller amounts. To illustrate:
Two positions are taken simultaneously, both at 1000$. Transaction fees= 1%
Our Revenue= 2 * (2%) * (1000$)= 40$
One performs 20% and the other one -20%.
Our Revenue= (2%) * (800$) + (2%) * (1200$)= 16$ + 24$= 40$.
The winning investor ends up paying 20+16+24= 60$ instead of 44$, which makes a 36% increase in overall transaction fees and owns 1200-60=1140$ instead of 1156$, or 98.6% (1.4% loss) of what he would have earned without the model.
The losing investor ends up paying 20$ instead of 36$ which makes a 44% overall decrease in transaction fees. He owns 800-20=780$ instead of 764$, or 102.1% (2.1% earning) of what he would have earned without the model.
In this example, the model hurts the winners less than it benefits the winner.
Henry said on March 17, 2010
Jonas, computing the average brokerage fee on the market requires empirical data that are collected by brokers themselves. They would obviously not publish it. But to give you an idea, you can check these transaction fees charts from the brokers Swissquote (Swiss) and Zonebourse (French):
http://www.swissquote.ch/index/index_services_f.html
http://www.zonebourse.com/mods_a/pdf/Tarifs_BD_ZONEBOURSE.pdf
One couldn’t say that the average return of the markets are predictable. Different theories exist. Some pretend that you can predict the future based on historical data, some others that you can rely on information published by companies, and some others that you must obtain the information from the inside. Some say that the markets are controlled by the majors and that they will dramatically fall sooner or later. Some capitalists bring up the fact that growth can never stop and that markets simply cannot break. Some use technical analysis, some fundamental analysis, some the japanese candlesticks, some create time series on risk management softwares. Without mentionning the hidden determinants of human behaviour… SO What’s predictable in the end? And with what interval of confidence? I don’t know.
If markets take off and work again, and that there exists indeed a population of medium risk aversion profile investors, the model should work.