Great article in this month’s Bloomberg Markets Magazine on High Frequency Trading (HFT) which, according to Tabb Group, has increased from almost nothing to account for 61% of US stock market activity and 70% of individual trades since 2005. The SEC announced it is to review this area of trading last month but the focus of the article is what this means in reality from a broker dealer view.
Working with Ancero, Bloomberg has compiled a list of the best brokers using their global trade execution prices between July 2008 and June 2009. This is anannual review, and shows the impact of HFT upon the brokerage community as Goldman Sachs is the clear winner for brokers handling over $25 billion in trades annually.
According to Ancerno, Goldman is the closest at getting orders filled nearest to the price when the order is received whilst JPMorgan, last year’s #1, has fallen to joint fourth with Barclays Capital, behind Bank of America Merrill Lynch and Morgan Stanley.
The World’s Best Brokers Loss in basis points*
#1 Goldman Sachs -27.5
#2 BoA Merrill Lynch -32.7
#3 Morgan Stanley -33.6
#4= JPMorgan Chase -34.2
#4= Barclays Capital -34.2
#6 Investment Technology Group (ITG) -35.6
#7 UBS -36.3
#8 Deutsche Bank -38.8
#9 Citigroup -39.7
#10 Credit Suisse -41.3
A basis point is 0.01%, so Goldman Sachs' 27.5 = 0.275% or just over quarter of one percent. This means that, for a client who placed an order for 50,000 shares at $10 each with Goldman Sachs, they would get the shares for an average price of $10.275 cents; whilst Bank of America would fill the order at an average price of $10.327 cents; Morgan Stanley at $10.342 cents; and so on.
[Figures represent the difference between the executed stock price and the price when the order was placed for brokerage clients in the four quarters ended on June 30th 2009, according to Ancerno]
Why have Goldmans won?
According to Roger Freeman, an analyst covering brokerage houses and exchanges at Barclays plc, because “they have the most developed and advanced electronic systems” and “can get some of the fastest execution times on trades.”
This is why latency is so critical and illustrates the point well.
For example, per day, the US exchanges averaged trading of 10.4 billion shares daily during the year to June 30 2009, with institutional investors globally paying $28.2 billion in trading commissions compared to $30.7 billion in 2008 and $26 billion in 2007.
Interestingly, the basis point difference varies quite widely by region:
Loss in basis points* USA Europe Asia
Goldman Sachs -25.4 -32.6 -24.0
JPMorgan Chase -34.1 -35.8 -31.1
This may be a reflection of the early state of low latency in Europe at that time, and demonstrates why Chi-X has succeeded so fast.