They are also very powerful rules, with the capability of significant impact; I have seen uncertainties in their application virtually shut down insurance company general account private placements for a period of time. And given that the value of the assets in IRAs and retirement plans are equal to some 85% of the value of publicly traded securities in the U.S., they will serve as a growing shadow regulation on a large piece of the U.S. economy over time.

It is also why 408b2 is also so intriguing to me. It has the potential for some pretty serious ramifications. Though we get lost in the detail of timely meeting the new disclosure requirements, the real impact will occur once the dust settles, and when we have to deal with all manner of prohibited transactions- arising either from failure to properly disclose compensation or from what is revealed by the disclosure itself.

Interestingly enough, one of the most serious of the impacts of 408b2 promises to arise from application of Code section 4975, not from ERISA Section 406 to which 408b2 is connected.  For those not familiar with it, Code section 4975 is the Tax Code's corollary of the prohibited transaction rules under ERISA's Title 1. By a quirky operation of the ERISA Reorganization Plan, the DOL has the authority to issue the regs by which the prohibited transaction tax under 4975 is operated (except for computation of the tax itself), and thus 408b2 acts to interpret 4975 as well. There are, by the way, important distinctions between 4975 and 406 which will come into play (for example, 4975 does not apply to 403(b) plans, but will apply to non-ERISA IRAs).

4975 imposes a tax on the prohibited transaction. It is not a penalty (though there is a penalty in 4975 for failure to timely pay the initial tax and correct the transaction), like under ERISA Section 406. It is a tax on a transaction. Period. There is strict liability for the taxes.  Once the prohibited transaction occurs, the tax liability attaches, and there is a duty to report and pay that tax. The IRS has no ability to waive that tax-unlike the prohibited transaction penalty under ERISA, which grants the DOL the ability to waive penalties-other than through the issuance of a formal prohibited transaction exemption by the DOL.  

This is even if the transaction occurred without any intent to engage in the transaction, or even if it occurred when engaged in doing what is ultimately in the best interests of the plan and participants. 

The 408b2 regs have an important PT exemption for the responsible plan fiduciary (which also applies to the 4975 tax) under certain conditions, should there be a failure of disclosure, but this exemption does not run to the service provider who fails to make a timely disclosure. And even then the exemption only runs to the disclosure itself, not the actual prohibited transaction which may be disclosed under 408b2.

The rubber has not yet begun to hit the road.... 

The DOL has saved those regs from chopped liver status, today making it clear in Question 2 to FAB 2012-2 that the 408b-2 relief for 403(b) plans also is extended to the 404a-5 requirements for said plans, on the same terms and conditions.

I, like many others, struggle with the notion of advocating for the ability to terminate a 403(b) retirement plan.  However, any retirement plan needs to be able to be sanely unwound -often for a number of critical, exigent reasons unrelated to the plan itself.  We now find ourselves in a growing number of difficult circumstances where the right solution is to terminate a 403(b) plan, but find ourselves unable to do so.

The IRS’s ongoing position is that an individual 403(b)(7) custodial account cannot be distributed from a 403(b) plan upon its termination, while a “fully paid” annuity contract is permitted to be distributed- as it will not deem the custodial account to be an annuity for these purposes.  This has the practical effect of preventing termination of any 403(b) plan which is funded with individual custodial accounts. As time wears on, and this position begins to get long in the tooth, its “unworkableness” becomes more and more apparent as it causes difficulties beyond the actual plan itself. 

I have never fully understood the legal concerns that the IRS has with treating the custodial account as a deemed annuity and thus a distributable asset from the plan, but will fully admit that this arises from my perspectives on these contracts bred by my particular experience with them.

Could it be that the IRS is uncomfortable with the notion of treating the custodial account as an annuity means treating the custodian contract of a plan as being, itself, a plan asset?  If so, this concern is eminently understandable. It seems counterintuitive, at best, that a part of the fundamental structure of a plan be considered a plan asset.  It makes much better sense that the underlying securities (or the cash generated from their sales) held by the custodian is the asset that can be distributed; and that its not really possible to distribute a portion of the structure of the plan. An individual custodial account cannot possibly be a financial instrument that can be recognized as an “in-kind” plan asset: its part of the plan itself.

Like many things 403(b), however, these sorts of issues are novel and of first impression when applied to such plans, requiring that we return to some very basic concepts.

I would hope that the following analysis would be useful.

Lets start with the distribution of an annuity contract from a 401(a) plan.  The tax regulations have long recognized that an annuity contract is a financial instrument (as opposed to cash) which can be distributed from a 401(a) plan (see 1.401(a)-20, Q&A 2); that it can have an account balance or a cash value (see 1.401(a)31, Q&A 17 which requires a qualified plan distributed annuity to allow a direct rollover of funds from the annuity contract to another plan or IRA); and the IRS has issued a number PLRs to that effect (see, for example PLR 200951039, which describes a variable annuity contract issued from a plan).

So it becomes a relatively straightforward proposition to permit the distribution of an individual annuity contract from a terminating 403(b) plan (putting aside for now the issue of what is a “fully paid” annuity contract). Even though it can be seen as constituting a part of the actual structure of a plan, its distribution is accorded no different legal or tax effect than is granted to annuity contracts distributed from a 401(a) plan.  And just as a qualified plan distributed annuity continues to be subject to 401(a) (as administered by the insurance company, see for example 1.401(a)-20 Q&A 17), the distributed annuity 403(b) contract continues to be subject to the terms of 403(b) (see Rev. Rul. 2011-7). 

But what of individually owned 403(b) custodial accounts which, like the annuity contracts they are deemed to be by the Code, seem to be an integral part of the plan’s structure?  Could it also be comfortably considered a financial instrument capable of being distributed by a 403(b) plan on termination in the same manner as an annuity?  

I invite consideration of the following: 

-The individually owned custodial account is a financial instrument, a legally binding contract which is signed by the participant (not the plan) and the custodian, like the individual annuity contract. It has specific terms and conditions on the exercise of the rights by the participant/owner over the underlying securities it holds. For example, the custodial accounts limit the distribution of its assets to those distributions permitted under 403(b)-11, and needed No-Act approval from the SEC to do so-as had the annuity contract. The legal right and title to those contracts under state law run to the participant (though also being subject to ERISA standards for ERISA plans), as does the rights under an individual annuity contract. It would be hard to argue that these are not ”bundle of rights” which constitutes property rights under traditional state property law analysis, as with an annuity contract. 

-The individual custodial account is fundamentally different than the traditional 401(a) custodial account, here with rights to the underlying assets themselves not only running directly to the participant, but the rights of participants can vary greatly within the same plan. 

-Given the state law property rights granted to the individual owner of the custodial accounts, and that the DOL uses ordinary notions of property law to determine whether or not something is a plan asset (see, for example AO 94-31-though the DOL has yet to take a position on this in the 403(b) custodial account context), and that annuity contracts are generally considered plan assets, it becomes hard to argue that the individually owned custodial account is not also a plan asset. And just because it is a plan asset which holds other plan assets (like annuity contracts with separate accounts), this does not disqualify its own plan asset "status."  I suspect that, if the custodial account were abused by a participant or vendor in some sort of way, the DOL would be the first to acknowledge that it is a plan asset.

-The custodial account can, and has historically has been able to, legally exist and operate independently from the existence of a plan and a plan sponsor, with the custodian acting in the same manner as an insurer under a 403(b)  annuity contract which is not related to a plan. 

-This financial instrument is recognized as a security by the SEC. The SEC’s own position recognizes the interests embodied in the 403(b) custodial account as a security owned by the participant, which further lends weight to the argument that it is an asset which can be distributed.  In Release 33-6188, it stated these interests are securities, though “as a matter of administrative practice...the staff does not require such interests to be registered."  This is not a reference to the underlying securities held by the custodial account, as those interests are specifically required to be registered. 

Admittedly, this presents a one sided view, but it is one which is both legitimate and could form the basis of permitting the distribution of the individual custodial account upon plan termination. Supporting this from a logistical view of things is that there are already a significant number of such custodial accounts described above in the marketplace which exist outside a plan. These came to be by virtue of 90-24 transfers occurring before 9/24/2007. Note that I do not address-nor advocate-the application of this approach to distributions from ongoing plans. There are a number of other complications which arise from that circumstance. Note, too, that I suggest that this analysis is only possible under 403(b) (and not under 401(a)) because of the specific language of 403(b)(7).

I would hope that this approach, recognizing the individual custodial account as the financial instrument it is, would make some sense to our regulators in deciding upon its termination distribution. 

"ERISA Account" is really not a defined term-anywhere-which is a problem in and of itself. So, for purposes of this piece, I refer to it as either an asset pool, or identifiable credits, that can be used at a plan's direction for plan purposes.   It is interesting what happens when you take time to discuss things openly with other professionals; a lot gets flushed out.  One of the important things we discussed was the differences in ERISA Accounts depending on the funding source. Working the details of the ERISA Account really involves working the details of the underlying investment product.  

Just take just one iteration of how ERISA Accounts can be funded, 12b-1 fees from mutual funds, for example. Though plan assets are used to purchase investment company shares, the actual assets within the mutual fund are not considered plan assets because of very specific ERISA rules (the shares owned by the plan are actually the plan assets; the underlying investments of the mutual fund itself are not). The mutual fund is governed by a board of directors, which is responsible for adopting any 12b-1 program that the fund has, and the program determines whether or not such funds are even payable to a plan to fund an ERISA Account. Any 12b-1 fees are paid by the mutual fund itself, not by the investment manager of the fund or the rep selling/advising-on the fund (though those parties may, themselves, be able to create ERISA Accounts if done properly, but not from mutual fund assets).  Though these 12b-1 fees may be considered indirect compensation for 408b-2 purposes, and even though they may be subtracted from the fund's value in determining the daily Net Asset Value of the shares held by the plan, the payment of the 12b-1 fee itself would not, generally be considered the payment of a plan asset.

One of the issues discussed was whether, if the plan had the legal right to that payment, whether that payment is a plan asset (or is the right to the payment an asset?).

If the 12b-1 payment is made to the plan's trust, it is a plan asset. But what if, instead, it is paid directly to the TPA, who uses it to offset administrative costs under its contract with the plan? Is it a plan asset? What about the advisor, or the plan's investment manager (particularly if it also manages the underlying mutual fund?).

This is just one example. Even within mutual funds, there are a variety of ways to handle the ERISA Account, and just wait until we discuss annuity contracts and collective trusts.  Determination of the plan asset status is just one of the issues (albeit a critical one) for any iteration of an ERISA Accounts, and we understand there is a pending DOL Advisory Opinion request somehow dealing with this sort of issue.  But then there are the questions of its accounting and treatment under the plan, the allocation issues, the forfeiture account issues, and even whether or not "definitely determinable benefit" becomes an issue if too much discretion on allocation is granted under the plan document. 

The only thing really clear is that there are a number of issues to resolve and, like prohibited transactions, the answers to them will lie in the specifics of the design of the ERISA Account; what type of entity funds it (and its relationship to fiduciaries and other parties in interest); how it is used; how it is terminated; what kind of contract commitment is involved; how is it documented and substantiated; how does it work in 403(b);  and others.

Whew. Here we go again......

I will be speaking on a panel addressing this topic with Mark Iwry, Jeff Turner and Wally Lloyd at the ABA Section of Taxation's Employee Benefits Committee general session on May 12 in DC. 

This means that financial service firms have a huge economic stake in not just making the initial disclosure, but also in making sure that permanent compliance procedures are established, implemented and audited on a routine basis. Firms should also be taking a closer look at their myriad of revenues streams to make sure that not only are they reported correctly, but that there is a Prohibited Transaction Exemption that otherwise permits the revenue. 

A firm, then, is faced with a quandary. "ERISA compliance" staffs (to the extent a firm has this function) have generally had the function to answer technical questions and provide procedural guidance to operation staff and clients. There really has been no enforcement "teeth." The only part of these financial firms that really have had to have serious "control" functions have been the securities law compliance staff. It is typically only these staffs which have the expertise necessary to establish, implement, audit, enforce and report regulatory requirements on the firms'  substantive business. And it appears that it is the securities compliance staff to whom firms are turning to institutionalize ongoing 408b-2 compliance.

I had the pleasure of being on a panel with Gigi Fuhry and Jim Downing at the National Society of Compliance Professionals meeting in Chicago on " Integrating ERISA Compliance Into the Securities Compliance Program," where we outlined the framework of an ERISA Compliance program that can built into the securities compliance practice. The NSCP itself now has training modules which are designed for the Compliance professionals who now find themselves confronted with ERISA.  I've linked to a compliance list the three of us developed.

Welcome, securities compliance, to our world!

One of the fundamental issues the proposed regulation raises is one which is beyond Treasury's control:  what is the fiduciary's exposure to the potential future insolvency of an insurer when choosing a QLAC provider? Absent a federal insurance system like the FDIC or the PBGC which ultimately guarantees this risk-which is likely to be a number of years off, if even possible-does this means that it will never be prudent for a fiduciary to purchase a QLAC, or any annuity, under which a single insurance company insures the risk?

Many of us have been seeking legislative and regulatory solutions to this for a number of years, to little avail. Having thought about this much, and having blogged on it on the issue on more than one occasion, I have come to the conclusion that it is unfair to expect the DOL  (and likely well beyond its mandate) to develop much more of a standard beyond what it has already published. If you read its "annuity safe harbor" standard closely,  you will see that it is comprehensive and describes the fiduciary standard well. But there still exists a queasiness about how to apply it.

I believe that the issue, however, is not as intractable as it seems, and the answer does not lie in further regulation. It lies in becoming more comfortable with the idea of risk and the commercial pooling of interests.  Fiduciaries should be able to comfortably assess an insurer without either becoming insurance experts or- at least on the narrow issue of insolvency- having to rely on an expert's prognostication as to whether an insurer will be around decades from now.

The real issue, as made starkly clear by the SCOTUS in the oral arguments on heath care reform, is that there appears to be a serious lack of understanding in the legal, judicial and investing communities of the nature of risk and of the commercial pooling of interest.  This lack of understanding seems to be underpinned by a reluctance to accept that we, collectively as a society, share certain risks that can only be managed at a more global level.  

There are a few key (but not so obvious)  concepts, that have existed for centuries that the law (and financial advisors) should recognize, and upon which fiduciaries should be able to rely:

1.  Risk is a natural part of life.
2. Our individual risks often are inter-related. Thus people can, and have, well managed these risks by pooling interests with those who have similar risks.  
3. Managing this pooling is complicated and requires specialized knowledge which has been successfully accomplished commercially (which also provides a level of financial accountability which structurally may be missing in government based programs).  
4. Financial "scale" is necessary to make risk pooling commercially work.
5. Such "scale"  and complexity makes it impossible for any single policyholder to protect itself against fraud and mismanagement of the pool.
6.  Collectively acting through government provides the regulatory scale necessary to protect citizenry from insurance (pooling) mismanagement and fraud, and the regulation of pooling is significant. Making sure commercial insurance is able to fulfill the pool's promise is an appropriate government function for protecting the well being of its citizenry.

In the end, the risk of insurer insolvency is a societal risk for which no fiduciary should be held accountable, as long as they are familiar the regulatory structure which is in place and uses it in making their decisions.

Regrettably, I believe much of the misunderstanding about pooled risk is reflected in deeply rooted political beliefs which was epitomized by the "Ownership Society" concept promoted by the Bush II administration. Though no one can deny the need for accountability, one should neither deny that longevity risk can really only be managed by acting in accordance with our common, and thus pooled, interests.

__________________

I cringed when I realized that we in the industry really have no one to blame but ourselves. 

When I first joined the insurance industry, the management consulting firms were working with insurance companies to address  insurance's pretty bad reputation. The term "insurance" had fallen into such ill-repute that even long term industry professionals became reluctant to say that word. I  sat through a number of industry meetings where I was amazed at the seeming repulsion at using it.  What grew from all of that was an easy way out: instead of referring to insurance policies, we would just instead refer to them as "products." Or, in the retirement world, "retirement products." Even today, look hard to see how often you see reference to a "group annuity contract "or "insurance policy" in the materials where your plan may have purchased one. It has to be disclosed by state regulation, but you'll only find it in the smallest print possible and in the most innocuous way. You are not going to see it highlighted in the midst of the marketing material, to be sure.

At first I railed against this practice. I had just come in from a manufacturing company where real, tangible product was being produced. Insurance, I insisted, was not cereal. Establishing the terms of an insurance policy is NOT "manufacturing," as the industry started to call it. It was, at best, annoying, and I believed a bit misleading.

I believe strongly in the commercial pooling of interests; it is one of those areas where good social policy and good business practices meet for the betterment of all. I 'm afraid that, on more than one occasion, I did get on my soapbox to claim that we had nothing to be ashamed of by being in the "insurance business", as long as we did it right. It is, after all, the selling of unique knowledge. 

Over time, however, I fell prey to the same practice. You may note in my "elevator speech" description of my law practice, I discuss "retirement products and services." But now all of that is coming home to roost.

The commercial pooling of interest is a complicated matter. Just think about the number of actuaries an insurance company may have (Lincoln, at its height, had 130 of them, and a formal actuarial recruitment, training and rotation program-complete with a sort of hierarchy that only an actuary can comprehend and appreciate) and the seemingly mind-numbing work they do. Even the recent QLAC regulations required the use of an actuary, and some of the terms are difficult, at best. The complications, the risks, and the reliance as a society we have on this pooling is also the reason it is so heavily regulated, in ways no box of cereal is.

It is all about providing financial services. No tangible product is made which can be handled and sold. Instead, we are talking about providing money and knowledge to policyholders under certain defined circumstances.

In the retirement world, for example, I have often referred to distributed annuity contracts as being "in kind" distributions, as it simplifies matters greatly. But as we delve deeper into the regulation of these things, we do ourselves a great disservice-and are likely to to get the regulations wrong- if we view insurance as a product as opposed to a package of financial services; it is about the way we appropriately regulate the pooling of our financial interests.

Insurance is not a cell phone. It is not even broccoli.

We are very fortunate, however, that the IRS staff responsible for implementing and enforcing those regulations recognized this failing early on, and has often sought to administratively address some of the more difficult concerns that have cropped up through this transition period.

Most recently, Monika Templeman (Director of Employee Plans Examinations for the IRS's TE/GE Division) announced in February an interim program to address an audit problem related to the IRS's delay in updating the EPCRS program to reflect 403(b) document rule changes, and the stalled 403(b) prototype program.

This program, according to IRS staff, will not be provided in formal, published guidance, but will be implemented by the field staff during audit until such time as EPCRS and the prototype program are finalized. It will helpful to the practitioner to be familiar with how this works, because it is relief that operates within limitations of the IRS's formal audit closing structure.

If an employer is found not to have adopted formal 403(b) plan document in a timely manner, the IRS may choose (presumably under non-abusive circumstances) to enter into a type of agreement where sanctions more resembling those available under the “Voluntary Corrections Program” (VCP) will apply, rather than those under the more expensive “Closing Agreement Program” (CAP). This should help encourage employers who still have not adopted a formal program to act as quickly as possible to adopt a formal written plan document, albiet late.

If a "document error" (as that term is used under EPCRS) has been found,  the employer may be given the option of amending the plan prospectively and fixing the past error as an operational error (either as a self correction, where applicable, or using the VCP structure); or to instead commit to adopting a prototype program (when issued) and follow the rules for the remedial amendment period under that program. It is important to note that the IRS will follow up with the employer should the "prototype choice" be selected, and one should give it serious thought before making this particular choice.

What the IRS CAN'T do under favorable terms because of the legal structure of the audit program (though some practitioners apparently insist on doing so) is to permit the retroactive amendment of the plan on audit. Insisting on this fix will cause the higher CAP sanctions to be applied.

On another note, it has come to my attention that my prior blog related to the "mis-marketing" of the 403(b) SPARK standards by some parties could have been read to imply that SPARK itself has been involved in this practice.  SPARK has not represented or suggested that its best practices are intended to serve as a DOL disclosure solution.  It was never my intent to lead any one to this conclusion, as SPARK is not engaged in such a practice. I do apologize for any misunderstanding that this may have caused.

The IRS also imposed new rules for exchanging contracts between vendors in the same plan. These new rules include the infamous "Information Sharing Agreement" (or "ISA") requirement, that conditions the exchange on the employer entering into an agreement with the issuer of the contract into which the funds will be exchanged. Under the ISA, the employer and the new issuer agree to, "from time to time in the future," provide each other Information necessary for the new contract, or any other contract to which contributions have been made by the employer, to satisfy section 403(b). This includes information concerning the participant’s  employment  status, eligibility for a hardship, and ability take a loan (and in what amount),  as well as Information necessary to satisfy other tax requirements.

In response to these new data exchange requirements, a number of friends and colleagues from a variety of 403(b) vendors joined together and  worked diligently and effectively over several years following the publication of the regulations to establish information standards for the exchange of this data. They are known as the SPARK 403(b) Information Sharing Data Elements Best Practices. 

Now, however, there appears to be a number of parties that are touting these ISA standards as the solution to 408(b)2 and 404a-5-compliance; that the SPARK standards will provide a solution to these new DOL disclosure rules; or that adopting the SPARK Standard suggests capability to successfully implement the disclosure rules.

Don't be misled; nothing can be further from the truth. The SPARK standards address the basic contract information which is necessary for compliance with the ISA requirements. There is no fee data; there is no investment data; there is no description of services. Reference to the "SPARK Standard" as a solution to the DOL rules does present a very catchy marketing and sales hook, but one which is also very wrong.

As an aside, the ISA standards were well developed by dedicated professionals, many of whom are passionate about this business. They have been, in effect, a tremendously successful  "Skunk Works."   Much of the standards' current value derives from volunteers who have moved on to other positions within and without the original organizations which were committed to the process.  So sustainability now becomes an issue. The standards were also developed under the auspices of a small organization with very limited professional and administrative staff which-in addition to attempting to establish annuity data standards along side the 403(b) standards-is very much focused on establishing its own new "brand." Sustainability, structure and support  are necessary elements for the maintenance and further development of these standards as they mature. 

Ultimately, as I've suggested in the past, the answer may well lie in the formation of a consortium funded in large part by user fees; supported by a well established center such as a public university; and governed by a wide variety of stakeholders-not just vendors. This is not just a pipe dream, this is how a number of standard setting organizations and consortia are run with a degree of permanence. I admit here to a bias, as a number off us had proposed and developed just such a structure in the early days of the new regs, and a number of critical parties were interested.

It will be interesting to see what happens over time.

  __________________

Think twice, however, before attempting to purchase what I call a "retail," or "nonqualified" annuity for a qualified plan. Annuities which are designed to be purchased by individuals outside of qualified plans, or are designed for IRAs, may actually cause a number of difficulties for the 401(k) plan. They really should only be used in small, specialized arrangements where the plan sponsors are cognizant of the special challenges presented by these products.

These retail products are typically designed to be sold individually by insurance agents and registered reps to what is sometimes called the "high net worth" (HNW) market. The are often sold as part of complex estate planning efforts, or for business succession. What generally makes these things inappropriate for the typical 401(k)plan is that:

 -They are complex. Simple annuities can be complicated enough, but the terms and conditions for the sometimes exotic guarantees within the typical GMWB contract or GMAB contract can make your head spin. As a fiduciary, it will be difficult to explain these complications on a mass basis, and many of them are just inappropriate for the smaller account balance.

-They can be expensive. In part because they are so complex, and can take a bit of effort to fit into the financial scheme of the policyholder, their fees are can be a bit "salty."  The mortality and expense charges are generally well above those you will find in a 401(k) plan, and the commissions paid on these products can be significant.

-Lack of ERISA compliance support.  Financial service companies are interesting places. Products sold to individuals typically are on different systems, and administered by different staffs, than those sold to retirement plans.  Administratively, even if one comes to terms with the complexity and the expense, the insurance company may not be set up to provide the 408b(2) disclosures; the 404a-5 information; Schedule C or Schedule A information; and may not even have the suitable SSAE-16 opinion.

-Minimum premium. These retail products can have hefty minimum premium requirements which, if applied in the qualified plan context, could cause serious Benefits, Rights and Features discrimination issues.

 -Harris trust.  And, of course, our old friend "Harris Trust." Products sold to the individual marketplace never had to deal with the issues related to protecting the general account assets of an insurer from being considered "plan assets" subject to ERISA governance. Many of these retail products do have a "fixed fund" based upon the investment in an insurer's general account. If the terms related to that general account benefit are not designed properly, the insurer may have some challenges.

Should your client really want the retail product, the solution then is to purchase it as a part of a rollover into an IRA.

  __________________

If, on the other hand, the plan leased several computers from Office Depot for the exclusive use by the plan in its administration, and there was an ongoing maintenance contract also purchased, the arrangement becomes subject to the prohibited transaction rules. An unreasonable payment for that purchase becomes subject to the prohibited transaction rules-and Office Depot would be required to return the unreasonable compensation, and be subject to the prohibited transaction tax. It is subject to 408(b)(2).

In the first example, Office Depot is neither a "party in interest" under ERISA Section 3(14)(b), nor a "disqualified person" under the Code's PT corollary, 4975((e)(2)(B)-which defines a party in interest/disqualified person as a "person providing services to the plan."  This means that mere sale of the typewriter by a non-party in interest to the plan isn't going to be a prohibited transaction, even if the price is unreasonable.

In the second example, the lease and the maintenance contract are likely both to be seen as providing services to the plan, which then makes Office Depot a party in interest to the plan subject to the PT rules.  Any unreasonable compensation would need to be returned by Office Depot.

The purchase of typewriters and computers from an office supply store is not covered by the new 408(b))(2) regulations, which only covers specified transactions. But example 2 is still governed by 408(b)(2), generally.

So why do I even bother raising this? This whole area of party in interest status under ERISA was a hot topic a few years back in the Mertens and Harris Trust (not the Hancock case, but Smith Barney) Supreme Court cases, where the Court was deciding whether or not to allow ERISA lawsuits against non-fiduciary parties-in-interest. But its important now, once again, because it really does serve as a threshold issue to the new 408(b)(2) regs.

If a person is a not a party in interest, (here, if the party is not one providing services to a plan, or a subcontractor of one providing a service to a plan), 408(b)(2) will not apply, or it will not apply until you are providing a service. So lets consider a few things:

Lets say an insurance agent is selling a group annuity contract (or a QLAC, for that matter) to the plan. She sells the contract, receives a commission, and walks away from the plan. The insurance company now deals directly with the plan, and the agent is out of the picture (which may happen for a number of reasons), and is not a subcontractor of the insurance company in the provision of services. I would argue that the agent is not a party in interest, so that 408(b)(2) would not apply to the payment of trail commissions to her. Schedule A reporting, however, would still need to be done. And unless she is consider a subcontractor of the insurance company for the services it provides, the insurance company would not disclose that compensation under 408(b)(2). In reality, most sales folks stay involved in a plan when receiving commissions, and do provide services (which triggers 408(b)(2)), but not always. 

Lets say that a promoter of a private placement or a hedge fund approaches a plan to sell the plan interests in a partnership or the fund. The sale is successful, the commission is paid, and the promoter walks away-the partnership or the fund now deals directly with the plan.  I would argue that the promoter is not a party in interest. You can sell, as I have said in the past, 'til the cows come home (or at least until your "sales" becomes "service", which is, under many circumstances though, inseparable-either as a CSP or a subcontractor).

And here's a much tougher one: suppose a marketing company related to a mutual fund complex approaches an advisory firm offering to pay $50,000 to sponsor a golf outing for the firm's advisors. The $50,000 payment is not related to, or based upon any level of production the mutual funds receive from the advisory firms clients.  It would appear that, under these circumstances, the marketing firm is not a party in interest which would need to report, and (even it were otherwise a CSP to a plan) this may not be indirect compensation that the advisory firm or the marketing firm would need to report to a responsible plan fiduciary, either (though there would be a potential PT issue, outside of 408b2, if the firm promoted this arrangement to influence its advisors). This type of arrangement, by the way, would have had to have been reported under the original iteration of the 408(b)(2) regs.

What this discussion points out is something that has been lost in the massive 408(b)(2) discussions in the market: this is about prohibited transactions, and prohibited transactions have ALWAYS been, uneasily, very fact specific. A slight shift in the facts will often determine whether or not a prohibited transaction has occurred.

So it is around what I call the "edges" of 408(b)(2). Though the regs will apply to a significant part of very common transactions and relationships, there are a number of them where the answer is not so clear. And, as with other eccentric prohibited transaction matters, a close look at the particular facts will be determinative.

Discretion being the better part of valor,  we should take seriously DOL's caution that the regs will be read broadly, and error should be made on the side of disclosure. But sometimes the rules will just not apply, and in circumsatnces that may be surprising.

 __________________

So it should have been no surprise when some very smart colleagues (whom I hold in high regard) were wondering about my enthusiasm about Treasury's newly released guidance on annuities (though there are a few others who do share this enthusiasm). After all, the rulings and the proposed regulation really seem minor in the grand scheme of things.  They did not resolve a number of issues, and almost seemed to raise more questions than they answered.

Sometimes, it seems, one can get too close to an idea and think that-of course-everyone sees what I see!

So, I've attached a graphic of how DC annutization works to help explain why the Treasury guidance is so fundamental and crucial to the next steps.  I invite you to take a close look at the chart in the first 4 pages of the patent application that Dan Herr and I filed in 2007, after actually working on it for a few years (it was assigned to my former employer, who has never used it; the patent was initially denied, never has been granted, and don't think it ever will (or can be)). It shows how your basic mutual fund 401(k) plan can serve as an annuity processing platform; how the distribution of annuities from the plan works; and why things like spousal consent and annuity starting date are so important to making it work.

The chart shows one way its possible to set up a QLAC with an automatic withdrawal program, using the plan's mutual fund, separate account, or even pooled investments which are then backed up with a QLAC or other lifetime guarantee. The chart is complex, in that the lifetime income demonstrated can be a set lifetime amount (such as a QLAC), variable annuitization, or even a GMWB.  It also shows how to do it within the plan itself, or to be distributed out of the plan, and how to do it using either an "in-plan" or "distributed" GMWB as well as a QLAC.

I was invited to Treasury's de-brief in DC the day the guidance was released, and it became clear to me then how this 6+ year old chart really puts the QLAC to great use (even for rank and file), and takes great advantage of the clarity provided in 2012-3.  Now can you understand my enthusiasm?

It ain't easy (yet), and it ain't pretty (yet). I caution that the description is in bureaucratic-speak, written by a patent lawyer in the way engineers are trained to do. But take some time on the charts, as they may help understand a few things about what may be going on with these things. 

The chart shows that Treasury actually answered key questions with its guidance. But my point is not that DC annuities are the "be all and end all" of retirement security, though they have an important place in making the system work right.  Nor is it that the insurance industry is the knight in shining armor (clearly, I know  its underbelly well) for which I am some apologist.  Its just that we now have the basic structure in place under which an important set of other operational and legal questions can be identified, asked and answered in an identifiable framework;  the argument-so to speak-has been framed. The rulings importantly  recognized that DC annuities will be treated as investments, with some strings attached to protect spousal rights; that there is a basic annuity starting date rule; that there is a forfeitability approach; and that Treasury is thinking about reporting and disclosure. Given this, we now know what even to ask.

Join us, by the way,  for a teleconference by the ABA's Joint Committee on Employee Benefits on March 1, where we will go over some of this stuff, in English....

The Department was persuaded by commenters on the interim final rule to exclude all or that part of a Code section 403(b) plan (hereafter “403(b) plan”) that consists exclusively of “frozen” contracts or accounts, as described in the Department’s Field Assistance Bulletins addressing the limited application of the annual reporting requirements to such contracts or accounts.  Plan sponsors and fiduciaries likely would be unable to comply with this rule because they often have no dealings with the relevant plan service providers and are unable to obtain information about these contracts and accounts.  Accordingly, paragraph (c)(1)(ii) of the final rule now provides that, in the case of a Code section 403(b) plan subject to Title I of ERISA, the “covered plan” would not include annuity contracts and custodial accounts described in section 403(b) of the Code with respect to which the plan sponsor ceased to have any obligation to make contributions (including employee salary reduction contributions) and in fact ceased making contributions to such contracts or accounts for periods before January 1, 2009.  Further, the contract or account has to have been issued to a current or former employee before January 1, 2009; all the rights and benefits under the contract or account have to be legally enforceable against the insurer or custodian by the individual owner of the contract or account without any involvement by the employer; and such individual owner has to be fully vested in the contract or account. 

Cloning 408(b)(2) or 404a-5 wasn't sufficient, nor was relying on those thick prospectuses that accompany variable 403(b) contracts. As valuable as the DOL disclosure schemes are in the employer-sponsored, institutional context, and as important as prospectuses are in protecting investor interests, they the lack simplicity -or obscure in volumes of other data- that which is particulalry important to the individual 403(b) participant. These plan particpants are not fiduciaires, and often do not have anyone to be able to collect and compare data on their behalf. They are sold investment products directly. So it really becomes a very simple issue for that school teacher or adminstrator: how much sales commissions is my investment generating; what services am I getting in return; is there a way for me to compare it all; and how can I reasonably access comparative data on the investemnts themselves?

There is the practical problem: whatever would be recommended also had to be doable by the 403(b) vendor providing the product, so disclosure was based upon information that was already being collected in some way by providers in the ERISA world. The committee also recognized the value of the comparatve format for investments under 404a-5 (and the significant investment being made in those dsclosures),  and took advantage of the benfits of that work by recommending their use here.

The Committee, with skillful drafting support and guidance from ASPPA's Deb Davis and with Craig Hoffman's ongoing involvement in our work,   put in serious time over the past 6 mohts to balance the useful with the doable. Along with Lisa and Melody, Aaron Friedman, Carol Gransee, Chris Guanciale,  Scott Betts  and Theresa Ward put together what I consider a pretty good piece of "engineering."  Having known engineers from from my days in Michigan, who have told me that successful engineering is really the art of successful compromise (as, they have said, you cannot have the fastest, strongest AND most fuel efficent vehicle in one; its always a balance between them all), this group can lay claim to a pretty good result.

We all recognize that this first effort was not perfiect, and pratcice will be a good teacher for us as we try to bring a new level of transaprency into play.  There is likely to be changes as we find out what we really did here. But its a pretty good start.

But there is a risk nowadays in even referencing these two grand ideas in today's political environment: instead of being viewed as the firm basis of how the vast majority of us quietly operate, they seem to have been outrageously hijacked by political extremists (such as of the libertarian/Ron Paul/ Tea Party sort-of which I am not so inclined) for some specific end.

In spite of all that, there is something well worth mentioning along these lines about the striking impact of the work we do, something we are not prone to see while working the fine minutiae of our chosen profession.

If we step back for a minute, we can see the extraordinary policy underlying 408(b)(2), the prohibited transaction rules and the exclusive benefit rules (which apply even to non-ERISA plans).  These rules seek to set aside and protect from others the individual wealth of those who accumulate benefits under these plans. It became pretty plain to me while reading Absolute Monarchs, by John Norwich (a history of the Catholic Popes, which really is a brief history of the absolute power of royalty as well as the church over individuals), where, historically, an individual's financial well being was wholly dependent on the whims of powers that be.

We now have something very odd in man's modern history. The value of the funds which are now protected for participants in retirement plans by the Code, ERISA or both approximates 85% of the value of publicly traded securities in the United States.  Though this seemingly huge amount is not yet adequate to establish broad retirement security, it is material enough to take note: these pooled funds are outside of the legal reach of the unaccountable "whims" of those who have something other than the best interests of the participants at heart.  Imagine that. A significant and growing portion of society's wealth is institutionally dedicated in funded pools to the individual's well being, which are difficult to access by an abusive use of power which has so often corrupted society-and jeopardized freedom and liberty-in the past. 

The only way this really works, however, is by things like 408(b)(2); by enforcement of the prohibited transaction rules; and by giving serious attention to the exclusive benefit rules. And all of this is dependent on what appears to be non-sensical minutiae upon which we daily work.

There is a reason I like doing what I do......

The first thing that comes to mind when looking at the regs for these purposes is the thing I noted in that previous blog: regardless of what you may think about 408b2 and the requirements now imposed by the rules, this reg has been craftfully drafted. The pieces fit together nicely, and complex issues with regard to investment products have been meaningfully addressed in as simple and direct manner as possible.There may be a few interpretaive issues that need to be resolved (which is to be expected), and 403(b) issues continue to be a serious challenge, but this is a fine piece of technical writing.

So it is with the "guaranteed account," "stable value fund" or "fixed account" within these group annuity contracts.  The 408(b)(2) pieces fit well together.

First, it may be helpful to read my piece on general accounts, and how they work. This can go a long way in understanding why 408b2 applies in the way it does. It is one of my favorites, because it includes a Dick Van Dyke clip from "Mary Poppins." That blog only generally addressed the 408b2 issue.

Next, the fiduciary needs to know whether or not the investment account is a contractual obligation of the insurer,  backed by the "general assets" of the insurer, or if it is part of a "separate account" (see my above noted blog on this). Confusion may arise from the variety of marketing names these funds may be called: the general account product will sometimes be called a "stable value fund" (typically because the crediting rate is set by use of an unrelated, third party index), which may be confused by a "stable value" mutual fund or collective trust interest which is offered under the annuity contract's variable investment separate accounts.

Once you know that fund is a fixed obligation of the insurer from its general account, you need to see what kind of Covered Service Provider (CSP) is the insurer.  It will not be considered an "A" type with regard to the fixed account, because it is not a fiduciary with regard to the management of the assets of the general account backing its contractual promise (a book could be written on this point, alone).  It may be a "B" type of CSP if, under the contract, the insurer is a "recordkeeper."  It will not typically be a "C"  type of CSP because, even though it is insurance, the insurer will not usually receive indirect compensation (as that term is defined by 408b2) related to that account (but keep in mind that these contracts may be part of an annuity where separate accounts are used-and thus indirect comp typically received-for which "C" status may occur). 

What needs to be disclosed?

-If the plan will only buy a group annuity contract with a fixed fund from the insurer, and the insurer will not provide any participant level recordkeeping services, then the insurer may not even be a CSP that will need to report under the new 408(b)(2) regs. The only complication will be the reporting of commissions. 

-If the insurer is a "B" CSP, then two things will need to be reported:  (i) a description of any compensation that will be charged directly against the account balance in connection with the acquisition, sale, transfer of, or withdrawal from the product, like surrender charges; and (ii)  description of any ongoing expenses such as mortality and expense charges or contract charges.

What will NOT need to be reported, however, is any "spread" between the crediting rate under the contract and the investment return on the insurers' general account, as the regs specifically exempt "operating expenses" from needing to be reported from a fixed account.

Didn't say it would be easy, but it all does fit together nicely.....

__________________

Any discussion on any tax issue addressed in this blog (including any attachments or links) is not intended to be used, and cannot be used, for purposes of avoiding penalties imposed under the United States Internal Revenue Code or promoting, marketing or recommending to another person any transaction or tax-related position addressed therein. Further, nothing contained herein is intended to provide legal advice, nor to create an attorney client relationship with any party.