Lightspeed Jacks Up Commissions
- Thread starter jeb9999
- Start date
So are any/all of IB/TR's internalized fills among the "95%" quoted earlier in the thread?
What most people miss when looking at internalization
is that there are several traders involved.
The customer sending a marketable order,
the internalizing B/D
and the trader(s) whose limit(s) is making up the NBBO.
If the B/D uses no information regarding the customer (how informed he is)
and has no view on the short term direction in the stock (unlikely)
when deciding to trade against the customer, then here is how the parties
are affected:
-- customer benefits via price improvement (typically ~ 0.01 c/share)
-- B/D benefits via jumping the queue (typically ~ 1c / share)
According to Tradeworx public commentary on the
SEC market structure concept release:
"Here is why it matters:
In a price-time priority market,
orders that are at the front of the queue experience the LEAST adverse selection,
and orders that are the back of the queue receive the MOST adverse selection.
This is obvious, because if you are the last one to buy on the bid,
that means the bid is about to become the new offer.
Conversely, if many people are behind you on the bid,
that means the bid is likely to hold after you trade.
Empirically there is a 1.7 cps difference in profitability for a posted
share that is first in line vs one which is last in line"
-- the trader making up the NBBO is robbed of the fill and the trader(s) with limits
in line behind him are robbed of the opportunity to advance in the queue
(total cost ~ 1c / share spread among them and felt via adverse selection)
If the B/D uses information regarding the customer or has the ability to
forecast short term direction, then the customer will expereince adverse selection
on their internalized fills. This cost will be higher for
better traders (informed traders) and when trading against a broker that is
particularly good at short term forecasts, and may very well overwhelm
the nominal price improvement received.
is that there are several traders involved.
The customer sending a marketable order,
the internalizing B/D
and the trader(s) whose limit(s) is making up the NBBO.
If the B/D uses no information regarding the customer (how informed he is)
and has no view on the short term direction in the stock (unlikely)
when deciding to trade against the customer, then here is how the parties
are affected:
-- customer benefits via price improvement (typically ~ 0.01 c/share)
-- B/D benefits via jumping the queue (typically ~ 1c / share)
According to Tradeworx public commentary on the
SEC market structure concept release:
"Here is why it matters:
In a price-time priority market,
orders that are at the front of the queue experience the LEAST adverse selection,
and orders that are the back of the queue receive the MOST adverse selection.
This is obvious, because if you are the last one to buy on the bid,
that means the bid is about to become the new offer.
Conversely, if many people are behind you on the bid,
that means the bid is likely to hold after you trade.
Empirically there is a 1.7 cps difference in profitability for a posted
share that is first in line vs one which is last in line"
-- the trader making up the NBBO is robbed of the fill and the trader(s) with limits
in line behind him are robbed of the opportunity to advance in the queue
(total cost ~ 1c / share spread among them and felt via adverse selection)
If the B/D uses information regarding the customer or has the ability to
forecast short term direction, then the customer will expereince adverse selection
on their internalized fills. This cost will be higher for
better traders (informed traders) and when trading against a broker that is
particularly good at short term forecasts, and may very well overwhelm
the nominal price improvement received.
Good post maler. Not to mention this practice disincentivezes other liquidity providers from posting bids and offers in the displayed market. Often times you get filled when you are wrong, because the "informed" trades are internalized or the flow is sold to hft firms leaving the displayed nbbo the LAST in line.
Any time a select few are able to do something forbidden to the rest of the trading community an edge is created. This is no different. The largest institutions have this edge as well as the political influence to keep it.
Adapt and overcome. This business has never been fair to small time traders and never will be. The most I expect the SEC to do is limit it to .01 and eliminate the sub penny price improvement.
Any time a select few are able to do something forbidden to the rest of the trading community an edge is created. This is no different. The largest institutions have this edge as well as the political influence to keep it.
Adapt and overcome. This business has never been fair to small time traders and never will be. The most I expect the SEC to do is limit it to .01 and eliminate the sub penny price improvement.
Internalization is a key part of many brokers - banks do this all day long. UBS has a system that matches orders from various clients. But I think instead of clients matched against each other, UBS acts as the middle man there and the trade is registered as OTC. Not sure the mechanics on the back end, but from a client's perspective, you order essentially jumps the queue.
Trade A post a limit order. UBS can't fill it. Sends this to the exchange. This limit order is sitting behind 100 orders.
Trade B sends a market order. UBS knows of the order for Trader A in the market. Sends a cancel to the market. Market order hits Trader A's order.
UBS charges both sides a small fee. If a partial fill, the left over gets sent to the exchange with best price/liqudity. The only thing I am not certain is how is this not slippage for the market order? But maybe clients are willing to go through UBS dark pool to get better execution fees/and not disclose large executions to the market.
One question this is how much of this is going on and what is the true price of a product you are trading. If 50% of actual trades happens in the dark pool, then you are probably not getting the right info.
Good thing exchanges are fighting against these since they are taking away executions from them. But how much can they fight against their own clients??
Btw, smaller brokerage do this as well. Peak6 in Chicago does this. I am not sure if this is also termed central hedging.
Trade A post a limit order. UBS can't fill it. Sends this to the exchange. This limit order is sitting behind 100 orders.
Trade B sends a market order. UBS knows of the order for Trader A in the market. Sends a cancel to the market. Market order hits Trader A's order.
UBS charges both sides a small fee. If a partial fill, the left over gets sent to the exchange with best price/liqudity. The only thing I am not certain is how is this not slippage for the market order? But maybe clients are willing to go through UBS dark pool to get better execution fees/and not disclose large executions to the market.
One question this is how much of this is going on and what is the true price of a product you are trading. If 50% of actual trades happens in the dark pool, then you are probably not getting the right info.
Good thing exchanges are fighting against these since they are taking away executions from them. But how much can they fight against their own clients??
Btw, smaller brokerage do this as well. Peak6 in Chicago does this. I am not sure if this is also termed central hedging.
I've been trying out Lightspeed with a demo account and have found their platform to be horribly buggy. The bid and the offer take turns freezing up, and sometimes they both freeze. Sometimes the whole platform freezes. Many times the connection is lost. I don't know why anybody would trade with this company.