Canadian Broker

[KCLarkin]

CIPF is okay, but have you looked at their financials?  I did actually look at them a few years back and my recollection is that their reserves were pretty modest.  It's not like CDIC or CMHC which are ultimately backstopped by the federal government.

Actually the CIPF is in pretty good shape. It has about half a billion in reserves. Maybe that does not sound like a big number but it's significant when you consider the actual payouts. In the past 10 years it had to pay out about 15 million for 4 broker failures. Since inception in 1969 it paid out about 69 million for 21 broker failures. In this light I think that 1/2 a billion is quite adequate.

 

The truth is that broker failures just aren't that expensive. It's really only client cash that's at risk in a failure and in some extreme cases some stuff around securities lending.

 

Yes. CIPF would be perfectly adequate for IB. You would need to burn through all of IB's reserves. Then CIPF. And with asset segregation, it is pretty hard to see how IB loses any significant client assets. IB invests its assets in 2 year U.S. treasuries.

[KCLarkin]

For small accounts like that, the big-boy benefits are modest and somewhat counterbalanced by integration of brokerage accounts with retail banking services.

 

$200k is actually a decent size. Don't have numbers for Canadian brokers but average account equity at TD and ETrade is $107k and 77k USD. Average account at IB is $200k.

[KCLarkin]

Thanks for the advice, but I won't hesitate to use a market order on WFC, BAC, RY, or CM.  They have that enormous bid-ask spread of 1 cent.  If I get screwed, it might be for $20 on a $100k trade, and I can tough that one out.

 

In May 6, 2010, Apple traded for $100,000 per share. P&G dropped 37%. Accenture traded for $0.01. That 1 cent bid-ask spread is an illusion.

[rb]

In May 6, 2010, Apple traded for $100,000 per share. P&G dropped 37%. Accenture traded for $0.01. That 1 cent bid-ask spread is an illusion.

I'll leave aside the flash crash stuff. I'll instead focus on the spread which some people seem to realize and you captured it perfectly. That spread is an illusion. Especially when you send a marked order. When you look at the spread it always looks like a 1 cent spread to the naked eye. But when a market order hits, the market dissapears for a few nanoseconds (not detectible to the naked eye) and you get a shitty fill. Those costs add up by the way.

 

Now some people may be ok with that, but I personally don't like bottom feeders picking my pocket by attaching themselves as leaches to my trades.

[StubbleJumper]

CIPF is okay, but have you looked at their financials?  I did actually look at them a few years back and my recollection is that their reserves were pretty modest.  It's not like CDIC or CMHC which are ultimately backstopped by the federal government.

Actually the CIPF is in pretty good shape. It has about half a billion in reserves. Maybe that does not sound like a big number but it's significant when you consider the actual payouts. In the past 10 years it had to pay out about 15 million for 4 broker failures. Since inception in 1969 it paid out about 69 million for 21 broker failures. In this light I think that 1/2 a billion is quite adequate.

 

The truth is that broker failures just aren't that expensive. It's really only client cash that's at risk in a failure and in some extreme cases some stuff around securities lending.

 

Yes. CIPF would be perfectly adequate for IB. You would need to burn through all of IB's reserves. Then CIPF. And with asset segregation, it is pretty hard to see how IB loses any significant client assets. IB invests its assets in 2 year U.S. treasuries.

 

 

What's IB's Canadian client base?  As a simple mental model, a loss of $50k for each of 10,000 clients would exhaust $500m of reserves...or 20,000 clients each losing $25k.  What's the likelihood of a systemic event where IB and some other brokerage are both trying to tap into the CIPF at the same time?

 

The scenario where there would be a failure would be some silly event in financial markets, or ridiculous financial malfeasance at IB>

[rb]

CIPF is okay, but have you looked at their financials?  I did actually look at them a few years back and my recollection is that their reserves were pretty modest.  It's not like CDIC or CMHC which are ultimately backstopped by the federal government.

Actually the CIPF is in pretty good shape. It has about half a billion in reserves. Maybe that does not sound like a big number but it's significant when you consider the actual payouts. In the past 10 years it had to pay out about 15 million for 4 broker failures. Since inception in 1969 it paid out about 69 million for 21 broker failures. In this light I think that 1/2 a billion is quite adequate.

 

The truth is that broker failures just aren't that expensive. It's really only client cash that's at risk in a failure and in some extreme cases some stuff around securities lending.

 

Yes. CIPF would be perfectly adequate for IB. You would need to burn through all of IB's reserves. Then CIPF. And with asset segregation, it is pretty hard to see how IB loses any significant client assets. IB invests its assets in 2 year U.S. treasuries.

 

 

What's IB's Canadian client base?  As a simple mental model, a loss of $50k for each of 10,000 clients would exhaust $500m of reserves...or 20,000 clients each losing $25k.  What's the likelihood of a systemic event where IB and some other brokerage are both trying to tap into the CIPF at the same time?

 

The scenario where there would be a failure would be some silly event in financial markets, or ridiculous financial malfeasance at IB>

Keep in mind that what we're talking about here is that only cash balances are at risk. So IB's average equity per account is 200k. If 10% of that is in cash you're talking 20K per account. 50k is unreasonable. Also if you're talking max loss, you're talking about a malfeasance event where IB managed to burn through all of their capital and somehow managed to squander all of its customers cash which basically impossible.

 

Then on top of that CIPF is supposed to run out of money because this should be an event of a magnitude that is 50x greater than anything that was ever experienced - as a reference MF Global cost CIPF about 2.5 million. Then on top of that Lloyd's becomes insolvent and defaults on the extra $2 million dollars per account coverage.

 

Yea, I won't say that the scenario is impossible because risk people would always conjure something up. But at that point your securities are worth zero anyway so it doesn't really matter whether IB is alive or not. Are you following the facts or are you making up your own?

[Cardboard]

One thing for sure, IB has a great and very smart leader. He learned from growing up in Eastern Europe unlike some: "That is why I vote Republican on every ballot."

 

Cardboard

[StubbleJumper]

CIPF is okay, but have you looked at their financials?  I did actually look at them a few years back and my recollection is that their reserves were pretty modest.  It's not like CDIC or CMHC which are ultimately backstopped by the federal government.

Actually the CIPF is in pretty good shape. It has about half a billion in reserves. Maybe that does not sound like a big number but it's significant when you consider the actual payouts. In the past 10 years it had to pay out about 15 million for 4 broker failures. Since inception in 1969 it paid out about 69 million for 21 broker failures. In this light I think that 1/2 a billion is quite adequate.

 

The truth is that broker failures just aren't that expensive. It's really only client cash that's at risk in a failure and in some extreme cases some stuff around securities lending.

 

Yes. CIPF would be perfectly adequate for IB. You would need to burn through all of IB's reserves. Then CIPF. And with asset segregation, it is pretty hard to see how IB loses any significant client assets. IB invests its assets in 2 year U.S. treasuries.

 

 

What's IB's Canadian client base?  As a simple mental model, a loss of $50k for each of 10,000 clients would exhaust $500m of reserves...or 20,000 clients each losing $25k.  What's the likelihood of a systemic event where IB and some other brokerage are both trying to tap into the CIPF at the same time?

 

The scenario where there would be a failure would be some silly event in financial markets, or ridiculous financial malfeasance at IB>

Keep in mind that what we're talking about here is that only cash balances are at risk. So IB's average equity per account is 200k. If 10% of that is in cash you're talking 20K per account. 50k is unreasonable. Also if you're talking max loss, you're talking about a malfeasance event where IB managed to burn through all of their capital and somehow managed to squander all of its customers cash which basically impossible.

 

Then on top of that CIPF is supposed to run out of money because this should be an event of a magnitude that is 50x greater than anything that was ever experienced - as a reference MF Global cost CIPF about 2.5 million. Then on top of that Lloyd's becomes insolvent and defaults on the extra $2 million dollars per account coverage.

 

Yea, I won't say that the scenario is impossible because risk people would always conjure something up. But at that point your securities are worth zero anyway so it doesn't really matter whether IB is alive or not. Are you following the facts or are you making up your own?

 

 

Am I following facts or making up my own?  What's that supposed to mean?

 

I asked a very basic question about the adequacy of CIPF reserves because they always struck me as pretty small in the context of the assets they are intended to protect, and you are suggesting that I am inventing facts?  Thanks a lot, buddy.

[Cardboard]

Am I wrong to think that it is not only cash but, all your assets that are segregated from your broker if you have a cash account (not margin)?

 

Therefore, cash accounts should have zero risk related to a broker's default. Maybe a delay in accessing your assets but, that is it?

 

Cardboard

[KCLarkin]

What's IB's Canadian client base?  As a simple mental model, a loss of $50k for each of 10,000 clients would exhaust $500m of reserves...or 20,000 clients each losing $25k.  What's the likelihood of a systemic event where IB and some other brokerage are both trying to tap into the CIPF at the same time?

 

IB has $4.5B in equity invested in 2 yr U.S. treasuries in its global brokerage business. You need to burn through that before you even need CIPF.

 

The CIPF might not have enough assets if one of the big banks failed. But it would have sufficient reserves to cover the failure of multiple small brokers (including IB).

 

And I don't know how IB could possibly lose $25K per client. The only real way for IB to lose money is for 1) Margin loans to go bad, 2) Collateral held is insufficient to cover losses, and 3) Unable to recover those loses through the courts. Keep in mind that IB monitors collateral in real-time, so you would need an event where you had great depression style crash but occurring in minutes instead of years.

 

I guess the other disaster scenario would be the U.S. government defaulting on 2 year treasuries. But in that case, the too big to fail banks won't be any safer.

[StubbleJumper]

What's IB's Canadian client base?  As a simple mental model, a loss of $50k for each of 10,000 clients would exhaust $500m of reserves...or 20,000 clients each losing $25k.  What's the likelihood of a systemic event where IB and some other brokerage are both trying to tap into the CIPF at the same time?

 

IB has $4.5B in equity invested in 2 yr U.S. treasuries in its global brokerage business. You need to burn through that before you even need CIPF.

 

The CIPF might not have enough assets if one of the big banks failed. But it would have sufficient reserves to cover the failure of multiple small brokers (including IB).

 

And I don't know how IB could possibly lose $25K per client. The only real way for IB to lose money is for 1) Margin loans to go bad, 2) Collateral held is insufficient to cover losses, and 3) Unable to recover those loses through the courts. Keep in mind that IB monitors collateral in real-time, so you would need an event where you had great depression style crash but occurring in minutes instead of years.

 

I guess the other disaster scenario would be the U.S. government defaulting on 2 year treasuries. But in that case, the too big to fail banks won't be any safer.

 

 

Yeah, that makes sense.  The $4.5b of treasuries strikes me as more useful than the Lloyds cover that was linked yesterday, which IIRC has a policy max of $150m.

 

When I looked at that CIPF fund (and even the Lloyds cover), my quick mental short-cut was to take the $400-500m of reserves and divide it by my account holdings, and the result was that it wouldn't cover many complete wipe-outs of people like me.  Then when you and RB suggest that it's mainly cash at risk, I mentally took the $400-500m and divided it by some of the dry-powder balances that I've had in the past and it provides a better result, but it still wouldn't cover many complete wipe-outs of cash balances held by people like me (and I'm not a rich guy!).  So, then I tried to take the opposite tack, and ask how many clients IB has, and divide the reserves by the possible size of the clientele to get a sense of what kind of an event would be required to cause a real problem.

 

I certainly heartily agree that the CIPF is small relative to the big banks' client base (would each of the big banks have 100,000 brokerage clients?).  The too big to fail status is much better protection for the banks' clients.

 

 

SJ

[rukawa]

The only problem I see with IB is that for penny stocks their fees can be higher. For penny stocks you are probably best off with TD or one of the big banks since they don't even charge you ECN fees, whereas Questrade and IB both do. Canadian exchanges often charge about  $0.0035 per share.

 

I generally only use limit orders. For IB does better execution just mean that limit orders will have a greater likelihood of filling than TD or Questrade?

[rb]

Am I wrong to think that it is not only cash but, all your assets that are segregated from your broker if you have a cash account (not margin)?

 

Therefore, cash accounts should have zero risk related to a broker's default. Maybe a delay in accessing your assets but, that is it?

 

Cardboard

yes

[rb]

CIPF is okay, but have you looked at their financials?  I did actually look at them a few years back and my recollection is that their reserves were pretty modest.  It's not like CDIC or CMHC which are ultimately backstopped by the federal government.

Actually the CIPF is in pretty good shape. It has about half a billion in reserves. Maybe that does not sound like a big number but it's significant when you consider the actual payouts. In the past 10 years it had to pay out about 15 million for 4 broker failures. Since inception in 1969 it paid out about 69 million for 21 broker failures. In this light I think that 1/2 a billion is quite adequate.

 

The truth is that broker failures just aren't that expensive. It's really only client cash that's at risk in a failure and in some extreme cases some stuff around securities lending.

 

Yes. CIPF would be perfectly adequate for IB. You would need to burn through all of IB's reserves. Then CIPF. And with asset segregation, it is pretty hard to see how IB loses any significant client assets. IB invests its assets in 2 year U.S. treasuries.

 

 

What's IB's Canadian client base?  As a simple mental model, a loss of $50k for each of 10,000 clients would exhaust $500m of reserves...or 20,000 clients each losing $25k.  What's the likelihood of a systemic event where IB and some other brokerage are both trying to tap into the CIPF at the same time?

 

The scenario where there would be a failure would be some silly event in financial markets, or ridiculous financial malfeasance at IB>

Keep in mind that what we're talking about here is that only cash balances are at risk. So IB's average equity per account is 200k. If 10% of that is in cash you're talking 20K per account. 50k is unreasonable. Also if you're talking max loss, you're talking about a malfeasance event where IB managed to burn through all of their capital and somehow managed to squander all of its customers cash which basically impossible.

 

Then on top of that CIPF is supposed to run out of money because this should be an event of a magnitude that is 50x greater than anything that was ever experienced - as a reference MF Global cost CIPF about 2.5 million. Then on top of that Lloyd's becomes insolvent and defaults on the extra $2 million dollars per account coverage.

 

Yea, I won't say that the scenario is impossible because risk people would always conjure something up. But at that point your securities are worth zero anyway so it doesn't really matter whether IB is alive or not. Are you following the facts or are you making up your own?

My snide remark about making your own facts was related to the fact that in the risk world you can always make up a scenario in which even the best capitalized and conservatively run insurance company goes bust. Nuclear explosions, etc. Generally something resembling the apocalypse.

 

However insurance is designed to deal with for lack of a better word (that is escaping me right now) "real world" situations. In that framework CIPF is well capitalized to deal even with the most extreme situations. In that framework total wipeouts of which you speak of don't happen. There are several reasons:

 

1. The business activities of the brokers do not predispose them to losses that large.

2. The regulator will wind up the broker well before it gets to that point.

3. There are several layers above that will absorb shortfall. The broker's reserves/capital, bondholder, uninsured client cash.

 

All of this is born out empirically in the loss history of CIPF which I gave in a previous message. So based on the loss events CIPF is likely to see, it is well capitalized to deal even with the most extreme ones. However it will likely fall short if the rapture should come to pass.

[bizaro86]

I'm mostly with IB but keep RBCDI accounts for penny stocks and convertible debentures, which are more expensive and not available on IB respectively. I've executed Norberts Gambit a number of times with simultaneous trades in my registered account.

 

I agree that the compliance could be bad, but is that actually a short position? If I have a settled long position by the time my sale needs to settle, isn't that just a sale of something I own?

 

Genuine question, this is way cheaper than what they charge for Fx, and I want to keep the account...

[StubbleJumper]

I'm mostly with IB but keep RBCDI accounts for penny stocks and convertible debentures, which are more expensive and not available on IB respectively. I've executed Norberts Gambit a number of times with simultaneous trades in my registered account.

 

I agree that the compliance could be bad, but is that actually a short position? If I have a settled long position by the time my sale needs to settle, isn't that just a sale of something I own?

 

Genuine question, this is way cheaper than what they charge for Fx, and I want to keep the account...

 

 

IMO, it's about the same as day-trading in your RRSP.  Day trading is not something that I do, but there are plenty of folks out there who have done a lot of it for a great many years. 

 

So, if you buy TD with T+3 settlement (or maybe now it's T+2?) and you sell TD a few hours later, you haven't actually created any phantom shares.  The only difference with a Norbert's Gambit is that you buy on one exchange and you sell on the other.

 

I'd be quite surprised to find that the brokers have not been in compliance for decades on this...

 

SJ

[rb]

I'm mostly with IB but keep RBCDI accounts for penny stocks and convertible debentures, which are more expensive and not available on IB respectively. I've executed Norberts Gambit a number of times with simultaneous trades in my registered account.

 

I agree that the compliance could be bad, but is that actually a short position? If I have a settled long position by the time my sale needs to settle, isn't that just a sale of something I own?

 

Genuine question, this is way cheaper than what they charge for Fx, and I want to keep the account...

 

 

IMO, it's about the same as day-trading in your RRSP.  Day trading is not something that I do, but there are plenty of folks out there who have done a lot of it for a great many years. 

 

So, if you buy TD with T+3 settlement (or maybe now it's T+2?) and you sell TD a few hours later, you haven't actually created any phantom shares.  The only difference with a Norbert's Gambit is that you buy on one exchange and you sell on the other.

 

I'd be quite surprised to find that the brokers have not been in compliance for decades on this...

 

SJ

Ok, so I need to issue somewhat of a retraction on the issues I've raised regarding naked shorts and so on regarding the gambit. The issue arise because stocks would have different CUSIPs for the Canadian and US listed shares. Making them different securities. This was true a few years ago when I last checked.

 

However, in preparing to respond to bizaro's post I went to check a few companies (RBC, Magna, Potash) and found that now they have the same CUSIPs and ISINs for both the US listed and Canadian listed common. In this case there is no problem - because they are the same security - and you can gambit away to your heart's content.

 

I should add that I don't know whether there has been an initiative to normalize the CUSIPs for cross listed securities and now all of them have matching CUSIPs and ISINs. So I would recommend to double check that the CUSIPs or ISINs match on your favourite gambit stock.

[KCLarkin]

So I tried my first currency conversion with IB yesterday. I just used a market order since currency markets are so liquid. The order went fine but the next day I got an email from IB:

 

We noticed that you have recently submitted Market Orders in your account(s) ********. Please see important information below regarding this order type.

Please note that a Market Order is an instruction to trade your order at any price available in the market, subject to any additional instructions for handling/simulating the particular order type you specified and other order conditions you specify when submitting your order. A Market Order is not guaranteed a specific trade price and may trade at an undesirable price. If you would like greater control over the trade prices you receive, please submit your order using a Limit Order, which is an instruction to place your order at or better than the specified limit price, or submit an algorithmic Market Order (IBALGO).

In accordance with our obligations as a broker, large Market Orders may be split into smaller orders, which will be traded over time. This is designed to reduce the impact of these large orders on the market, including the impact your order has on the market price.

Interactive Brokers

 

In other words: don't use Market Orders.

[NormR]

I'd like to thank everyone for their commentary on this thread.  It has been quite helpful!