TSP investors suspending contributions

Now here are the actual costs of trading:

IN 2007:

View attachment 5307


ANd then the same period in 2008: AFTER they imposed trading limits.



View attachment 5308

See much difference?

In all except the I fund, on both REAL TOTAL COSTS, and on "basis points" costs were WORSE, not better, after they imposed the trading limits.

In the I fund, the $8 million that was made in October could have been MUCH HIGHER had they allowed unlimited trades. That's a function of whether the markets ended up lower or higher on the overnight- something that the trading limits did not address. But TSP shareholders in general would have been MILLIONS OF DOLLARS better off had they allowed unlimited trades.

Now are you getting the picture?

Trading limits had nothing to do with costs.

It had everything to do with the TSP staff and the Thrift Board imposing THEIR investing management style opinions on the rest of us.

It was about CONTROL.

Nothing else.




 
Last edited:
Now,I don't blame Tracey Ray and Greg Long from doing what they thought at the time was in the best interest of shareholders.

I think what is clear now, though, is that it IS in the best interests of shareholders to allow unlimited trades. And it is now a good time to go back to the Thrift Board under the new Administration, and ask them to reconsider the limits that are in place.

Either by regulation, which is within the control of the Thrift Board and Staff of TSP, or by Congressional law change. That can be done too.

Either way, it needs to be looked at again, and , when the actual data is shown, the Board needs to change the policy back to more reasonable numbers.

Five trades per month instead of two?

Or maybe unlimited, with a $10 fee.

Either way would be a great improvement from where we are today.
 
I have actually increased my contributions since the market has self destructed. I figure that the lower the stock price the more shares I get. I do agree that later on when the stocks rebound, albeit 10 years from now, I can take the profit and reinvest it.
:cool:
 
James -

The new system may have been designed to handled unlimited IFT's by everyone, but I don't think it was intended to. Sort of like the "worst case scenario" capability built into the system.

Your argument is certainly persuasive after reviewing your statistics and graphs. From the numbers presented, the amount traded in the I Fund in October 2008 (for example) is significantly lower than the amount traded the previous October, when IFT's were unlimited. Assuming that the bulk of the October 2008 trading of the I Fund was more "institutional," vice participant-driven IFT activity, could that not affect the amount of the trading costs for that particular month? What other variables would affect the trading costs for a particular month for a particular fund? Is the current economic climate having an affect - would the numbers replicate themselves in a bull market? What other consequences of unlimited IFT's would be present - forcing the fund manager (is that the right term for the TSP) to maintain more cash on hand ?

Don't a great many private sector 401K's impose much harsher restrictions and/or fees on active trading? I realize that I'm no expert, but if what seems like a majority of 401K discourage active trading, can they all be wrong?

And if the TSP C, S, and I Funds are basically supposed to mirror their benchmark indexes, how does a high rate of IFT's affect the fund manager's ability to accurately mirror the index if daily changes to the fund are occurring through IFT's (is there even an affect?)?

The more I learn the more questions I have. Continue to stoke my curiosity at your leisure...
 
Too many questions all at once. Let's try and take them apart, one at a time.

Assuming that the bulk of the October 2008 trading of the I Fund was more "institutional," vice participant-driven IFT activity, could that not affect the amount of the trading costs for that particular month? What other variables would affect the trading costs for a particular month for a particular fund?
First, there is no such thing as "institutional" trading activity in TSP. No institutions own shares in TSP. It is only individual share holders who own shares.

Now, put this in your mind as well. Barclays is the agent which executes trades into and out of actual shares of stock. Remember, the only time Barclays gets involved is when there are MORE sellers than buyers, or visa versa. Otherwise, it's all handled in-house. If there are more buyers than sellers, (or reverse), then an order is placed at 2:30 each day to Barclays.

Barclays has two choices. Choice 1- they can meet the demand within their OWN system. Barclays keeps a slush fund of cash and shares within Barclays to address demand. Under SEC rules, there is a three-day clearning time frame, so Barclays typically keeps somewhere on the order of one billion dollars in both stock shares and liquid cash. If the fund managers at Barclays decide they can fill the order from within their own holdings, they can do that.

Note: TSP has no "fund managers". TSP only has order takers, no assets, and computer processing people. Barclays holds the cash, and shares.

Choice 2- IF the amount of the order exceeds the ability of Barclays to cover in house, they have to go out on the open market and buy or sell shares to cover the amount of orders placed. In this case, there are micro-fees associated. By Micro, I mean that yes, there is a "handling charge" associated with the brokers who are doing the buying and selling. If the order is, say, "BUY 24 million dollars of S&P 500", then there is a broker's charge for the exchange. The amount is TINY compared to the actual volume traded. You or I could not get the kind of rates and fees that Barclays gets on doing a 24 million dollar trade.

IN the month shown above (October 2008), Barclays traded 2.3 BILLION dollars in the S&P 500 ("C" fund) . That was spread over roughly 23 trading days. (23 executions of trades). It cost them a grand total of 223 THOUSAND dollars to execute all those trades. That works out to $1 in trading costs, for each $10,620 of shares traded during the month.

If I had $106,200 in a C fund trade, it would have cost the TSP a total of $10 to execute that trade.



The entire amount of variable trading costs rest within how Barclays executes the order.

IF they cover in-house, there is no expense at all.

IF they have to go out on the open market, then there is a small fee associated.

There is one more varible involved- that is "time".

They have to declare a share value each day. That's a whole differnet discussion, which doesn't need to be done here. Suffice it to say that the TSP has chosen to declare a "value" at 4 pm. each day.

But the trades don't occur at exactly 4 pm. The trades occur in the time between when Barclays get the order at 2:30 each day, and the market closing at 4 pm.

The Barclays fund managers get to decide whether to do it at 2:35 pm, or 3:58 p.m. Let say today Barclays gets a "sell order" of 100 million shares.

Now, in a falling market day, the fund manager would be smart, and execute the trade at 2:35 pm. Let's say he is right, and the market falls another 1 percent between 2:35 p.m. and 4 p.m.

The stocks were sold when the values were one percent higher then they were at the close. So there is "extra money" all of a sudden.

So, what happens to that windfall money?

Some of it is paid out in bonuses:
http://www.independent.co.uk/news/b...us-review-as-bank-seeks-uae-cash-1002752.html

Some of it is retained in the slush fund, to offset days when the fund manager misses the mark.

That brings us around to something called "Tracking Error" - something I'll cover in a future post.

Sometimes tracking error is in favor of shareholders, sometimes its against share holders.

The I fund is slightly different. The value is declared at 4 pm. But the trades are not actually executed until overnight- when the foreign markets open up and trades are made. The last market to open is FSTE, which opens at about 6 am. eastern time the following morning. It's crapshoot, because FTRIB doesn't set the value at 7 am the next morning. If they did, that would eliminate 95% of the varible costs of the I fund trades. We suggested that in the filings we did during the comment period of the trading restrictions. They didn't listen to us.

The "I" fund MADE money in October, 2008, because the value was set at 4 pm, and,. on a couple of days, the markets tanked or rose more than the fund manager expected. It turned out to be in the favor of shareholders by $8 million dollars.

But it could have gone the other way.

But still, the amount is tiny- minescule, compared to other investment vehicles. The way Barclays handles the slush fund ensures that. Barclays is very, very good at handling that slush fund.
 
Last edited:
Is the current economic climate having an affect - would the numbers replicate themselves in a bull market?

Yes, to the extent that there are larger than historically normal markets swings right now. If the swings were smaller, than the costs may be slightly higher, because there is less cushion for the Barclays fund manager to play with. Again, it's a crap shoot depending on how well the Barclays fund mangers project the market is going during each trading session, and whether or not they execute early or late.

What is more important is not whether the market is BULL or BEAR, but whether the size of the swing is large or small. A good fund manager can make more money either way when the swing it larger. The problem is making sure you are right more times than you are wrong.

You don't have to be right every time. they only count the money at the end of the month. You just have to be right more times than you are wrong. One bad big swing call, and you could blow a whole month of good little calls.

But again- and I don't know how many times I have to say this for it to sink in- we are talking a FRACTION of the costs as any private sector plan. TSP runs on operating costs of about 3 basis points a year. Private funds run on 60 to 80 basis points on a regular "norm". So if it ends up being 4 basis points instead of 3 basis points, we really should be very, very happy still. that's the part most TSP holders don't understand. It DOESN'T MATTER if they are off by a tiny fraction. Barclays guys are right more than they are wrong, and the larger the amount being traded, the better off they are at hitting the slush fund money in the right direction.

For October 2008, you see the number of just over $8 million in profit from the slush fund. In 2008, two-third fewer shares were traded than in October 2007.

If the same number of shares had been traded in 2008 as in 2007, that profit would have been three times that- or about $24 million.
 
Gibby: Don't a great many private sector 401K's impose much harsher restrictions and/or fees on active trading? I realize that I'm no expert, but if what seems like a majority of 401K discourage active trading, can they all be wrong?
Can they all be wrong? Yes.

They are discouraging active trading, as a reaction to an SEC rule that was issued after a Mutual fund scandal a few years back.

IN that case, one mutual fund started declaring "value" only once a week or so, and was funneling money to friends on the differences. There were two classes of share holders. those that got inside information, and those that didn't. those that did got huge profits when they traded after insiders tipped them off that the valuation was wrong, and that they were going to have a revaluation in a day or two. That allowed these big money insiders to rip off the common man. The insiders knew when things weren't right,and money was about to be made.

Of course, the SEC over-reacted, and laid down a rule that said all funds must declare a value once per day (a valid rule) AND that trading was to be discouraged (a NOT valid rule as it applied to the rest of the financial industry).

See http://www.filife.com/stories/mutual-fund-scandal-spawns-new-rules


These types of rules shouldn't apply when there is only ONE class of investor (Us).

And they shouldn't apply when the shares are valued every day. (Like they are with TSP funds.)

But the industry agreed to the SEC rules to prevent more harsh rules from being imposed, so here we are.

Those SEC rules, by the way, DON'T apply to the TSP. They only apply to mutual funds. TSP is NOT a mutual fund.
 
Gibby: And if the TSP C, S, and I Funds are basically supposed to mirror their benchmark indexes, how does a high rate of IFT's affect the fund manager's ability to accurately mirror the index if daily changes to the fund are occurring through IFT's (is there even an affect?)?
And THAT is the ONLY GOOD, VALID question that needs to be asked.

You are BINGO dead on with that question.

The TSP will be "successful" IF it is able to replicate the index that the fund is supposed to represent.

The goal is to be as close as possible to that index. If it hits exactly, then it has met it's objective, and is doing exactly what it is intended to do.

When TSP first started back in 1987, it was often OFF the mark. Wells Fargo was the first transfer agent, and missed the mark. That first year it was a struggle to match the market. here is the actual data from 1988.



As time went on, the tracking got better.

After a while, the transfer agent changed to Barclays, and Barclays has improved the tracking of the index ever since.

Last year, in the January to October time period, here is what the tracking was, with "unlimited trades":

[FONT=Arial,Helvetica,sans-serif]In 2007, the S&P 500, from January 2007 till September 2007, gained 9.13% The C fund gained 9.16% at the same time. They BEAT the GOAL by 0.03%. [/FONT]

[FONT=Arial,Helvetica,sans-serif]With unlimited trades. [/FONT]

[FONT=Arial,Helvetica,sans-serif]In 2007, the Wilshire 4500 index, from January 2007 till September 2007, gained 8.97%. The S fund gained 9.12% in that same time. They BEAT the GOAL by 0.15%. [/FONT]

[FONT=Arial,Helvetica,sans-serif]With unlimited trades.[/FONT]

[FONT=Arial,Helvetica,sans-serif]And in 2007, the EFAE International Index, from January 2007 till September 2007, gained an impressive 13.15%. [/FONT]
[FONT=Arial,Helvetica,sans-serif]The I fund, during that same period, gained 13.37%, beating the goal by an impressive 0.22%.

[/FONT][FONT=Arial,Helvetica,sans-serif]With unlimited trades.[/FONT]
 
And this year, now that they have LIMITED TRADES TO TWO PER MONTH- \
Here is the tracking data, January through October 2008:
View attachment 5315

Each of the funds has slightly beating the index it represents, in a time when we've had a MAJOR market contraction. the funds would have done better than that, in my opinion, had unlimited trades been in effect. Of course, we'll never know now, because the Thrift Boad took that option away.

In my opinion, "success" is defined when the funds meet or beat the indexes. If they can do that, then there should never be a restriction on trades.

Barclays has shown they are able to do that. They did it last year without limits, and they are doing it now with limits. In short, TSP is a success when they can meet the index. Anything above that is gravy.

Please digest that for a while, and then feel free to ask any more questions-

But try and keep the number of questions per post to one or two- it makes it easier for us to research and answer that way.

thanks.




 
BRAVO!!!!!!! BRAVO!!!!!!! now you get me pissed off all over again. James you are the man
 
Trading limits had nothing to do with costs.

It had everything to do with the TSP staff and the Thrift Board imposing THEIR investing management style opinions on the rest of us.

It was about CONTROL.

Nothing else.

I think we need to get 60 Minutes over to James's house.​
 
James -

If the costs associated with unlimited IFT's are miniscule, and a fraction of what private plans incur for similar activity, how do the costs change (assuming unlimited IFT's were in effect) if more TSP participants were actively trading?

I would think at some point as more participants conduct frequent IFT's the transactional costs would increase, and the ability of the fund manager/TSP/Barclays (whoever) to accurately mirror the benchmark indexes would be impacted. If the costs were low when less than 1% of TSP participants were actively trading, what happens when, say, 5%, 10% or 25% actively trade?

At what point is there too much play in the system where it produces negative consequences to the bulk of the TSP investors who are more long-term oriented (i.e., B&A's)?
 
First, there is no such thing as "institutional" trading activity in TSP. No institutions own shares in TSP. It is only individual share holders who own shares.

What I meant by "institutional" trading activity was the management of the fund to mirror the index, in accordance with more regular trading activity balance the fund to match the index, while taking into account contributions and redemptions. The normal trading activity excluding active trading.
 
James - At what point is there too much play in the system where it produces negative consequences to the bulk of the TSP investors who are more long-term oriented (i.e., B&A's)?

The only point where there needs to be any concern whatsoever, is if it begins to affect returns BELOW what the index is. I.E.- if the amount of trading causes the fund to NOT be able to mirror the index.

As shown above- we are a long ways away from that point. If Barclays is able to properly manage matching the index, then there is "success" in the plan. All indications are that Barclays has been able to match the fund indexes, and then some. As you can see in times of larger variations in the market (EITHER up or down), then Barclays has been effective in exceeding the Index they represent.

My arguement would be that in a free society, we should be able to exercise that right, right up until the point where they are not able to meet that goal. THEN, and only THEN, should any type of a charge to trade be considered.

My mathmatics tells me that we won't reach that at anywhere near the levels I would anticipate this freedom would bring.

And even if the charge to trade was, say, $5 per trade, that would generate sufficient funds that we would never reach that point.

If the costs were low when less than 1% of TSP participants were actively trading, what happens when, say, 5%, 10% or 25% actively trade?

Let's say that 10% of the shareholders actively trade, four times per month.

That would be on the order of 290,000 people, trading four times per month, which would equal 1,160,000 trades taking place.

If there were a $5 charge per trade to do that, that would generate $5,800,000 in income for the TSP per month to execute those trades, or $69,600,000 per year.

Currently, the ENTIRE BUDGET for the administration of the TSP is less than $90 million. Adding in $69 million in income from $5 per trade fees would far exceed the costs involved in executing all those trades.

Currently, our TSP Board allows Barclays to engage in Secutrities lending. That should concern you more. Securities lending is lettling others borrow YOUR SHARES to short the market. According the current November issue of the Meeting Minutes, Securities trading brings in some $90 million in income to the TSP annually.

The Thrift Board's reaction to that figure:

Here is a quote from the meeting minutes of November, 2008:

"Mr. Fink then asked whether the $90 million generated from securities lending was significant.
Ms. Ray noted that it is a few basis points, and Mr. Petrick
commented that is it the size of the Agency's budget."


So, Tracey Ray thinks that $90 million is "a few basis points". And she is absolutely correct. It's a few basis points.

Overall, the TSP costs are less than 4 basis points per year.

We are beating the indexes by "several basis points" now, and then some.

So there is NO REASON to limit trades at this point.

There is NO REASON to limit trades, even if 10% of the people were doing it. We STILL would be able to beat the index performance, if past mathmatical history is any indication of future performance.


My 2 cents...
 
What I meant by "institutional" trading activity was the management of the fund to mirror the index, in accordance with more regular trading activity balance the fund to match the index, while taking into account contributions and redemptions. The normal trading activity excluding active trading.

There is no "institutional activity" in this sense, with one small exception.

If a component of the index changes, then there is a mandatory change in allocation. Indexes don't change components very often. The S&P500 changed 23 companies between June 1 and October 20th this year, due to massive bankputcies (Lehman Brothers, Bear Stears, WAMU are just three of the 23 changed in the period of massive failures of large corporations. See http://seekingalpha.com/article/100912-nasdaq-comstock-added-to-key-s-p-indexes

In those kind of cases, when an index changes it's market basket, there is activity.

But short of that, "institutional activity", as you define it, doesn't exist.

And regular contributions currently far exceed redemptions, and most likely will for the foreseeable future.

There are more and more people being added to the rolls as contributors- far fewer are taking money out. Over the next decade, the number of redeemers will increase as baby boomers retire. However, the military number of contributors is expected to continue to grow significantly, and the number of new federal employees should remain pretty constant. Right now between 1.5 and 2 billion in new money is added each month as inflow.

The only growth area of "normal trading" recently, has been the introduction of the L funds, which rebalance every day. Tracy Ray claims that L fund rebalancing is miniscule. I believe it is a little more than that, but still it is not statistically signficant at this point.

Now, if 25% of the people held their money in L funds, then the volume could rise.

But even so, I don't think that amount of money moving daily is going to cause the funds to not be able to match the indexes they represent. Once again, Barclays has it down to a science at this point.

At what point is there too much play in the system where it produces negative consequences to the bulk of the TSP investors who are more long-term oriented (i.e., B&A's)?
If Barclays no longer can match the index, that is the point at which we need to take a look at it.

Maybe a $5 per month charge to all L fund holders, to cover the costs it takes for THEM to move their money on a daily basis. Remember, L funders have their money moved for them each and every trading day of the month. At what point do I, as a C fund holder, have a valid grievance against the L fund holders, that their moving of money on a daily basis adversely affects my returns? I will tell you that the costs are so small, that I am happy to allow them to utilize the L funds, as we STILL are able to meet and beat the index. Yet L funds DO take my money to administer for them.

Yet you don't hear the Thrift Board complaining about that.

No, costs are not excessive. Either for administering the L funds, NOR for allowing those who wish to actively manage their own funds. Returns in TSP continue to beat the indexes. If it ever is significantly less than that, THEN is the point at which to look at costs saving measures, or trading fees.

Until then, forget it. Costs are too small to worry.
 
James, my sincere and most humble thanks for stating the case for active trading so eloquently.

As I’ve said before, I am a FERS disability retiree. I have to make my TSP account last for decades. I cannot work another job to make my funds stretch. I cannot leave my account in “buy and hold” mode and take a chance that it will make enough money to cover my withdrawals.

By using my IFTs carefully in 2008 I was able to cover most of my monthly payments to myself while avoiding any losses to my account. If I had been able to actively trade my account I would have been able to add to my account substantially. I know that for a fact because I kept track of the trades I would have made if I had been able.

OBGibby, the facts James has presented on this thread are simple, accurate and irrefutable. I just pray that someday the TSP board will pay attention to them. My financial survival depends on it. If you are a federal worker, someday yours might too.

Lady
 

OBGibby, the facts James has presented on this thread are simple, accurate and irrefutable. I just pray that someday the TSP board will pay attention to them. My financial survival depends on it. If you are a federal worker, someday yours might too.


Lady

I agree with Lady James, very well said, and it's funny Lady that you mentioned "If you are a federal worker", because my read is that he isn't a fed.

CB
 
Back
Top