Naked Access

The SEC has just opened a comment period on the possibility of restricting naked access.  Naked access occurs when a broker sends a client order along to a market without doing any checks on that order.  Three examples of problematic naked access:

• the order might (erroneously) be larger than the market could handle;
• the order might require the client to commit more capital than they have; or,
  
• the order might, if filled, cause a client to exceed risk limits set by the broker.

The first problem is serious: it can disrupt the market until the offending order is found and people are called notifying them of the error.  The second and third problems are more subtle yet may also be nasty.  Such situations should never occur: uncollateralized and/or unlimited loans with no approval put brokers/custodians at great risk.  If you want to trade, you should always have enough money for margin (at least) and be within risk bounds you agreed to.  Why then do brokers not check for these violations?

Brokers do check for such violations.  However, talking to a few brokers suggests some only check at end-of-day.  Therefore, intraday violations are not found.  If a client enters and exits a gargantuan position intraday, the broker may never know the risk they were underwriting.  The lack of capital commitment also attracts manipulative traders.  Anecdotal evidence suggests some manipulative traders uses these loopholes.

Why would a broker offer naked access?  Speed.  Market participants compete over periods of microseconds to get the best price.  I think that is good (both from experience and theory) and much better to slower markets where friends got better execution than strangers.

Checking orders will slow them down.  That speed reduction could make checked orders less competitive.  However, if such checks applied to all orders, relative speed advantages should be unchanged.  The only change in competition might come if certain traders did not need to check their orders against risk limits.  Is that unfair?  Or is that just comparative advantage at work?  I'm not sure.
. . .
Regarding my preceding post: I am not of the Austrian school of thought.  I believe in unfettered markets, but I believe all theories and models are imperfect.  As George E.P. box said: "All models are wrong, but some models are useful."  However, when we compromise between models or theories, we should be open about what we are doing and why.