Comments Regarding SEC Proposed Rule:
Amendments to Rules Governing Pricing of Mutual Fund Shares
File No. S7-27-03
I am an individual investor whose primary investment of choice is mutual funds. I have invested in funds for two decades, three if one includes money market mutual funds. Over that period of time there have been many changes in the mutual fund industry and in thrift/retirement plans. I offer some observations on the SEC's proposed "hard close" rule from that perspective.
The purpose of the "hard close" rule is to preclude some investors from taking advantage of material information after the price of a fund's share has been set (at closing). If one investor can purchase shares at an old price, say, after there is a late evening announcement that is relatively certain to raise the NAV of the fund over the next day, then that investor benefits from that information while the other fund investors do not. I strongly support the goal of the proposed regulation.
The questions are whether a "hard close" is the only way to achieve this goal, and whether the costs outweigh the benefits. For the reasons outlined below, I believe the answer to the former is yes, and to the latter that the proposal is positive on balance.
2. Alternative of time-stamping is insufficient
Some commentators have suggested time-stamping as an alternate solution. This alternative appears in the SEC proposal as well. I believe this suffers from two fatal flaws.
In essence, this alternate proposal extends the group of trusted order handlers from a minimal set (the funds' transfer agents and registered clearing agencies) to third party intermediaries. Unfortunately, third parties, even subject to SEC regulation, have shown themselves to be untrustworthy. For example, several Prudential brokers in Boston have been charged by the Commonwealth of Massachusetts with submitting late trades for clients, even going so far as to hide the clients' identities1.
My personal experience has shown that brokers do not take the existing regulations seriously. Two years ago, I submitted an order through a brokerage arm of one of the largest financial institutions in this country at 4:15PM ET, intending that the order be executed the next business day. Instead, the brokerage aggregated my after-close order with others and sent that to the fund company. Thus my transaction was executed as a "late trade". The brokerage informed me that such aggregation is not uncommon. While fifteen minutes may appear to be just a small transgression, I see in this an industry contempt for rules that does not engender trust.
The alternate proposal of time-stamping by third parties has a second flaw. The proposal is advertised as a "technological fix." A time-stamping system is to be built that is tamper-proof. Consider the current discussion about electronic voting - many industry experts are skeptical about the ability to build a tamper-proof system2. An order-stamping system shares several attributes with a voting system, including tallying, data storage, data transmission, and desirability of an audit trail. Neither a system to tally votes nor a system to tally orders will be failsafe.
Finally, the goal of the SEC proposal is prophylactic - to protect against violations - not remedial. Thus, the fact that one might be able to track down a violation after the fact and compel restitution is not sufficient.
3. Cost vs. Benefit
Do the costs outweigh the benefits? The majority of objections to a "hard close" rule focus on the fact that third party intermediaries will have to set an earlier deadline for trades. How much of a problem is this?
3.1 Little if any benefit to placing orders at close of market
An investor who places an order earlier in the day trades with less information than a trader that places an order minutes before market closing. However, both people receive the same fair value price for the fund. This is to be differentiated from the original problem of late trading, in which the late trader effectively gets yesterday's price (i.e. a price that is mismatched to the value of the underlying assets when the trade is executed).
There is a psychological factor at play here. People feel that they are at an advantage if they trade knowing how the market did. One fallacy of this impression is that even if the fund share price can be reasonably anticipated near close of market, this says nothing about the next day's pricing. If an investor defers a purchase because the investor expects the share price to rise the day of the trade, that investor is betting that it will not rise more the next day. This is not predictable. On the other hand, if the investor makes a purchase because of an expectation that the share price will go down that day, the investor is betting that the fund will rebound the next day. Again, there is no way to predict this. The investor may feel better about the purchase, but doesn't know how that investment will perform. The same reasoning applies to sales of shares.
Another fallacy is investors' belief that they can accurately predict the movement in a fund's share price based on market movement. While this is substantially true for index funds, it is much less certain for actively managed funds. Ironically, the long lag that the SEC permits a fund in reporting its holdings makes it less likely that an investor can estimate the movement of a fund's price based on how the market or particular companies perform.
Like many people, I do try to trade near the end of the day. Objective observation over the years shows that decisions I made based on market movement were right about as many times as they were wrong. I experienced both types of problems described above. Sometimes after purchasing on a down day, the share price continued to fall. Other times, the share price for the day was up even though the market was down. Of course, this balanced out with times where my decision was "correct", i.e. the share price moved in the anticipated (hoped for, actually) direction.
3.2 Impact on fund supermarkets; direct investment alternative
Despite the futility of timing the market, many investors want to attempt it via mutual funds. These investors are free to purchase shares directly with the fund companies. If purchased that way, then they can trade right up to market closing. For funds that are priced at times other than 4PM (e.g. Fidelity's Select Portfolios®, Rydex Dynamic Funds), investors must already go directly to the fund in order to get a non-4PM price. It was not that long ago that if one went to a fund supermarket, one was required to submit orders significantly earlier than 4PM3. Supermarkets thrived despite this limitation. Investors who preferred convenience would trade through a supermarket; those who cared more about last minute trades would trade through the funds.
Investors have also learned that they may not be able to execute a same-day exchange across mutual fund families. Brokers do not know the exact pricing of a fund to be sold, and thus cannot submit a same day order to sell a holding in one fund and purchase shares of a fund in a different family. This is another example of compromises that investors have already demonstrated willingness to accept, in order to partake of the convenience of fund supermarkets.
Schwab has written that one of the benefits of supermarkets is that they exert a downward pressure on fund expenses (through competition). One can easily argue the opposite. Schwab charges around 40 basis points for participation in its OneSource program4. Many funds do not participate because this would raise the cost of their funds; others choose to pass the cost on to their customers in the form of higher operating expenses. Yet investors still choose to invest through supermarkets, because of the convenience of one stop shopping.
Because investors have already demonstrated flexibility regarding benefits and limitations of fund supermarkets, I do not feel that supermarkets will unduly suffer as a result of the proposed "hard close" rule.
3.3 Impact on retirement plans
Investors in retirement plans (e.g. 401(k) plans) must use third party intermediaries. For them there is no option to invest directly with the fund. It is here that one hears the most objections.
3.3.1 Early order deadline need not be very early
Plan providers say that they need to perform certain checks before passing orders on to the underlying funds. These checks are rule-based, and can be performed without human intervention. This is a situation where a technological fix is appropriate. Modern systems, greater processing power (more hardware), better compliance packages can reduce the time required to process orders. The time required will never be reduced to zero, but it can be minimized.
Software could precompute eligibility requirements, in preparation for a rush of orders. The SEC could add another exception to its "hard close" rule to accommodate this: A plan provider would be permitted to submit orders on the basis of, say, 99% certainty that the order is consistent with the plan requirements. In the rare cases where the order turned out to be improper, the plan provider could submit a late order, along with adequate documentation about why the order was barred.
Such a system could still be gamed, but the risk to the plan provider would not be worth any benefit. To cheat the system, the plan provider might submit orders that it knew in advance were not permitted. Then, depending upon later market conditions, the provider would either submit a cancellation (based on the plan rules) or silently ignore the violation. Patterns of such abuse would be easy to spot (a relatively high frequency of cancelled orders), rendering such action too high risk for the plan provider.
The point is that the difference between a plan deadline and the market "hard close" deadline can be reduced, though not eliminated. The more valuable such reduction is to a company and/or its employees, the more they will be willing to pay for it. Plan providers can compete at different points on the price/speed curve.
3.3.2 Early deadline not significant issue (despite hue and cry)
There is a more fundamental question, though. Should plan participants even be concerned with intra-day fluctuations in the market?
Defined contribution plans are long term investments, intended for funding retirement. Participants lose much more by poor asset allocation than by small daily pricing changes. Through much of the 1980s, participants invested much too much in GICs and money market funds. Then in the 1990s, investing became fashionable, people rushed to equity funds that were excessively risky. I observed many people in high tech companies pour 100% of their retirement investments into technology funds. Watching the daily prices and ignoring the diversification and long term performance of retirement investments is missing the forest for the trees, and ultimately unproductive.
More and more, defined contribution plans are taking the place of defined benefit plans. So it makes sense to consider how defined benefit (traditional pension) plans invest. These plans invest prudently, with long term performance in mind. Even when investments are changed, the plans do not turn on a dime, based on how a market performs between 3 and 4PM. So it should not be necessary, or even advisable, to encourage this type of trading in defined contribution plans.
Defined contribution plans offer participants certain tax advantages. Along with these advantages come restrictions. People understand this tradeoff. There are penalties for withdrawals. Plans often disqualify an employee from contributing for a period of time after the employee ceases contributing. Investment options are usually limited to a small collection of funds that the employer (or its agent) has determined are suitable for retirement investing. Plans are not even required to offer daily trading. Narrowing the trading window by time of day instead of by particular days would simply be one of many restrictions imposed in exchange for the benefit of tax deferred investments for retirement.
3.4 Benefits outweigh costs
Because of all the reasons stated above, I feel that the impact (cost) of a "hard close" rule is small and can be minimized, while the benefits are significant, both monetarily (by excluding special deals for a few), and procedurally (by providing a bright line rule).
4. Short term redemption fees
Though redemption fees are not a part of this specific proposal, I feel I should mention here as well. All funds have trading costs; the more the fund trades, the higher the costs. It is unfortunate that trading costs are not called out in prospectuses, and I hope that at some later time the SEC will address this, for trading costs can be substantial.
Two factors affecting trading costs (aside from soft dollars) are turnover and volatility of cash flow. The former is obvious - the greater the turnover, the greater the volume of trades, and thus the greater the trading costs. The latter affects trading costs, because greater volatility in cash flow increases the churn of a fund (having to purchase stocks one day simply to sell them the next to cover redemptions). Alternatively, the fund may keep a greater cash cushion, increasing drag on fund performance.
Rapid trading by investors increases the volatility of the fund. This increases trading costs in a manner out of proportion to the actual magnitude of the dollars involved. Consequently it is not unreasonable to impose high redemption fees on short term trading. Ideally, one would have the funds estimate the impact of churning on trading costs, but then again, ideally one would have the funds document the trading costs in the prospectus. In the absence of documentation, a redemption fee of a few percent for a short period of time would not be unreasonable. As a long term investor, I would personally benefit such a fee, as it would be borne by those churning the fund, not I.
It may be advisable, however, to allow funds to opt out of this requirement. Such funds would be designed for traders. In exchange for opting out, the funds could be required to disclose, under the suitability section of the prospectus, that:
- The funds are designed for traders;
- The funds have higher operating expenses than comparable funds designed for long term investors (due to greater record keeping requirements, confirmation notices, etc.);
- The funds have higher trading costs than comparable funds designed for long term investors (due to greater volatility of cash flow);
- The funds are not well suited to long term investing.
High redemption fees have been suggested as a way to address stale pricing, where traders take advantage of funds that have priced their shares based on outdated foreign equity closing prices. The better solution is fair market pricing, as that directly addresses the pricing issue, rather than imposing high redemption fees, which merely discourage people from taking advantage of the mispriced funds.
Nevertheless, I support high redemption fees, because they discourage short term trading, while leaving mutual funds attractive for long term investors such as me.
I hope the SEC will consider some of the arguments I have presented, and regulate mutual funds in the best interest of the individual long term investor.
|1|| Secretary Galvin Files Complaint Against Prudential Securities, Inc. http://www.state.ma.us/sec/sct/sctpru/pruidx.htm
|2|| "Insider attacks are the most common, dangerous, and difficult to detect of all security violations." D. Jefferson, A.V. Rubin, B. Simons, and D. Wagner, "A Security Analysis of the Secure Electronic Registration and Voting Experiment (SERVE), January 21, 2004 at 13. http://www.servesecurityreport.org/paper.pdf
See also Eric A. Fischer, "Election Reform and Electronic Voting Systems (DREs): Analysis of Security Issues", CRS Report for Congress, November 4, 2003 http://www.epic.org/privacy/voting/crsreport.pdf
|3|| As recently as five years ago, many brokers required orders to be placed by 2PM. Pettit, Davidson, and Lott, "Mutual Funds - Buying from Brokers versus Fund Companies", The Investment FAQ (December 28, 1998). http://invest-faq.com/articles/mfund-buy-broker.html
|4|| Harriet Johson Brackey, "Schwab fund `supermarket' turns dubious", The Miami Herald Herald.com, November 30, 2003. http://www.miami.com/mld/miamiherald/business/columnists/harriet_brackey/7373330.htm?template=contentModules/printstory.jsp