How Do You Know Your Retirement Plan Makes Sense? The Forensic Questions I Ask to Find Out
Most retirement plan reviews start with a benchmark, but I prefer starting with questions.
Who is being paid? How much? For what? Who is actually paying? How was the arrangement created? What has changed? What work is actually being performed? Who chose the providers? What incentives do they have? And would the employer make the same purchasing decision if the services were presented the way it purchases other professional services?
I have been investigating retirement plans since 2009. I started by examining publicly available Form 5500 filings, contacting plan sponsors about what I found, and trying to understand what was behind the numbers. Over time, I began keeping records of those conversations and what happened afterward. That database now contains more than 7,000 conversations with people responsible for retirement plans.
That experience has changed the way I look at a retirement plan.
A fee by itself doesn't tell me very much.
A percentage doesn't tell me very much.
A benchmark doesn't necessarily tell me very much.
Even a fiduciary designation, a clean audit, or a fee disclosure doesn't answer the most important questions.
I want to understand the economic relationship.
What is the plan buying?
What does it cost?
Who pays?
Who gets paid?
Why is it structured that way?
What work supports the compensation?
And does the arrangement still make sense?
I don't ask every question in every investigation. The questions depend on the plan, its providers, its history and what the evidence reveals.
But these are the kinds of questions I ask.
1. What is the plan actually paying?
1. How much did the plan pay each of its providers last year—including the advisor, recordkeeper, TPA and other service providers—and what was the total amount paid?
2. What percentage of plan assets did each fee represent, and what did those percentages translate to in actual dollars?
3. Are there payments that don't appear as an obvious advisory fee, including commissions, revenue sharing or other indirect compensation?
4. Who receives each form of compensation, and how are revenue-sharing payments generated and allocated?
5. Does the Form 5500 tell the same story as the employer's understanding of what it pays?
2. How is the advisor compensated?
6. How is the advisor's compensation determined—is it based on plan assets, participant count, services, time, a fixed dollar amount, or some combination of these?
7. Why was that compensation structure chosen, who proposed it, and was the fee negotiated?
8. If the fee is asset-based, what relationship does the amount of assets have to the work being performed?
9. How does the advisor's workload change when plan assets, participant count or plan complexity change, and does compensation change accordingly—or does compensation increase simply because assets have accumulated?
10. If the advisor performs essentially the same work next year, why should compensation increase simply because plan assets increased?
11. Does the compensation ultimately reflect work performed or assets accumulated?
3. Who is actually paying?
12. Who ultimately bears each cost—the employer, the plan or participants—and through what mechanism is the compensation collected?
13. Does the employer understand who is actually bearing the cost, and do participants know when their accounts are being charged?
14. Is the person making the purchasing decision the same person bearing the economic cost?
15. Are investment expenses being used to compensate providers, and if so, does the employer understand that arrangement?
16. Why was the current payment structure chosen, and was an alternative structure ever considered?
4. What are the advantages and disadvantages of employer-paid versus participant-paid fees?
I don't think one structure should automatically be treated as correct. The important question is whether the tradeoffs have actually been considered.
17. What are the economic advantages and disadvantages of having the employer pay versus having participants pay, including the tax consequences, impact on participant accounts and value received by participants?
18. What does the employer save by shifting an expense to participants, and what does the participant give up?
19. Is the participant receiving something directly in return for a fee deducted from the participant's account, or is the service primarily being purchased by the employer?
20. Would the employer make the same purchasing decision if it had to pay the expense directly?
21. Has anyone actually compared the two payment structures and quantified the economic difference?
22. Is the current payment structure the result of a deliberate decision, or simply the way the provider has always collected its compensation?
5. Where is the invoice?
23. Does the advisor, recordkeeper or TPA send an invoice, and does the employer receive a consolidated invoice for the services it purchases?
24. If there is no invoice, why not?
25. Does the employer receive a recurring, understandable statement of the actual dollar amount being paid, or must it calculate or reconstruct the cost from percentages and multiple disclosures?
26. Does the employer know what it paid during the previous quarter or other billing period?
27. Does the payment process create a regular opportunity to reconsider whether the relationship and its cost still make sense?
6. What should the invoice tell the employer?
28. Does the invoice state total compensation in dollars and identify each source of compensation, including direct fees, revenue sharing and other indirect compensation?
29. Does it distinguish compensation paid directly by the employer from compensation deducted from participant accounts?
30. Does it describe the services actually delivered during the period being billed?
31. Does it distinguish recurring services from one-time services?
32. Does it allow the employer to compare the compensation with the work actually performed?
33. Does it provide enough information for the employer to question the fee and negotiate it?
An invoice isn't simply another disclosure.
It creates a purchasing event.
7. What services are actually being provided?
34. What exactly does each provider do, where do their responsibilities overlap, and which services are actually being performed by each provider?
35. Which services are required, which are optional, and which were actually delivered during the past year?
36. What work did the advisor actually perform—including meetings, participant assistance, investment analysis, plan design, fiduciary support, compliance and education—and how much time and expertise did that work require?
37. How much participant contact actually occurred, including participant meetings and calls?
38. What did the advisor do that the other providers didn't?
39. Are there services being paid for that nobody uses?
8. Does the compensation correspond to the work?
40. What work justifies the current compensation?
41. What would another qualified provider charge for comparable work?
42. Did the amount of work increase when compensation increased, and did compensation decline when the workload declined?
43. Did the plan become simpler or more complex without a corresponding change in the fee?
44. Is the fee based on something that actually drives the provider's workload?
45. Does the compensation bear a reasonable relationship to the services being provided?
9. What happened over time?
A single year's fee tells only part of the story.
46. How have the plan's assets, participant count, services, workload, providers, plan design and compensation changed over the past five and ten years?
47. When compensation changed, how much did it change, what caused the change, and did the change occur before or after the underlying change in services or circumstances?
48. Did compensation change when plan assets or participant count changed, and did it change in the opposite direction when those measures declined?
49. Did compensation continue after a provider stopped actively servicing the plan, the responsible person left, or the advisor died?
50. Did the fee ever go down?
51. When the workload declined, did compensation decline?
This is where a historical examination can reveal things a single year's benchmark cannot.
10. Who made the purchasing decision?
52. Who hired the advisor, when was the decision made, and what did the decision-maker know and believe they were buying at the time?
53. What providers, prices and compensation structures were considered, and was the fee negotiated?
54. Were employer-paid and participant-paid compensation structures discussed?
55. Did the employer understand that the advisory fee was negotiable and that the advisor was optional?
56. What happened after the original decision-maker left, and does the person responsible for the relationship today know what the advisor is paid and what the advisor actually does?
11. How much does the relationship influence the decision?
57. What pre-existing personal, family or business relationship does the decision-maker have with the advisor, and how long has that relationship existed?
58. Does the sponsor describe the advisor as valuable or say the advisor “does a lot” without being able to explain the work or compensation?
59. Is preserving a long-standing relationship making it harder to objectively evaluate the economics?
60. Has anyone separated the question “Is the advisor valuable?” from “Is the advisor's compensation reasonable?”
61. Would the employer still select the advisor if the existing relationship didn't already exist?
62. Would the employer choose the advisor today at today's price?
12. What incentives exist between providers?
63. What financial relationships and incentives exist among the advisor, recordkeeper, TPA and other providers, including compensation paid directly or indirectly through another provider?
64. Who benefits when plan assets increase, who benefits when the existing arrangement remains in place, and who loses compensation if a provider is replaced?
65. Does the advisor's compensation depend on assets remaining on a particular platform?
66. Does anyone have an incentive to question the arrangement?
67. Does anyone have an incentive not to?
Who is actually positioned to question advisor compensation?
68. Does the recordkeeper or TPA receive business through retirement-plan advisors, and how important are those advisor relationships to their businesses?
69. If a recordkeeper or TPA depends on advisors for business, what incentive does it have to tell an employer that its advisor is being paid too much?
70. Would a provider be likely to recommend replacing an advisor who brings it business?
71. Who within the existing provider structure is financially independent enough to scrutinize the advisor's compensation?
72. If the advisor's compensation is reduced, could that affect another provider's relationship with the advisor?
73. Who is actually positioned to tell the employer that the advisor's compensation doesn't make sense?
13. What happens when you question the arrangement?
74. Does the provider welcome the question and provide the underlying data and evidence?
75. Does the provider explain the compensation and the services that correspond to it?
76. Does the provider offer a lower fee or propose a different compensation structure?
77. Does the provider explain the advantages and disadvantages of employer-paid versus participant-paid fees?
78. Does the provider explain why an asset-based fee is preferable and what additional work corresponds to additional compensation?
79. Does the provider encourage the employer to compare alternatives?
80. Does the provider point to a benchmark instead of explaining the economics of the particular arrangement?
81. Does the provider point to fiduciary status instead of addressing the compensation?
82. Does the provider make it easier or harder for the employer to investigate?
14. What does the Form 5500 reveal?
This is where my own investigation began.
83. Who is reported as a service provider, how much compensation is reported, and what appears on lines 8f and 8g?
84. What has changed from year to year, including unusual increases, unexplained decreases and changes in provider names?
85. Does compensation change when the plan's circumstances change?
86. Does the filing contradict what the employer thinks it pays?
87. Does the filing reveal compensation the sponsor didn't know about?
88. Does the historical filing tell a different story from the current one?
89. Does it reveal provider relationships or compensation patterns that would be difficult to see from a single year's disclosure?
15. How were the investments selected and evaluated?
90. Why was each investment option selected, who selected it, what alternatives were available, and which alternatives were actually considered?
91. What plan-specific characteristics were considered in making the investment decision?
92. When was each investment and the overall investment lineup last formally reviewed?
93. What documentation supports the decision to retain each investment?
94. How frequently is the investment lineup formally reviewed?
95. If the investment lineup has remained substantially unchanged for years, what substantive analysis supports continuing to use the same investments?
Target-date funds
96. Why was the target-date fund family selected, and why has it remained the plan's target-date family?
97. What other target-date families were available through the recordkeeper's platform when it was selected, and which were actually considered?
98. How were the alternatives compared in terms of glide paths, underlying investments, expenses and historical performance?
99. What characteristics of the plan's participant population were considered when evaluating the glide path, including participant ages, contribution rates, turnover, withdrawal patterns and expected retirement behavior?
100. What characteristics of the target-date fund make its glide path appropriate for this particular participant population?
101. Has the target-date series been formally reevaluated since its initial selection, when was the most recent documented review, and what alternatives were considered?
102. What evidence supported the decision to retain the existing target-date family?
103. Has the target-date fund's relative performance been evaluated over an appropriate period and compared with funds using different glide paths and investment approaches?
104. Were more conservative and more aggressive glide paths considered?
105. If the fund has underperformed relevant alternatives, what analysis supports continuing to use it?
Share classes and investment costs
106. Is the plan using the lowest-cost share class reasonably available, and if not, why not?
107. What services or benefits justify using the higher-cost share class?
108. Are additional expenses compensating a provider for services that would otherwise be paid separately?
109. Has the cost of available share classes been compared and revisited as lower-cost share classes became available?
Investment review and advisor services
110. How frequently does the advisor conduct a documented investment review, and what does that review actually examine?
111. Does the advisor evaluate target-date funds separately from other core investment options and compare investments with appropriate alternatives?
112. Does the advisor document why an investment should be retained?
113. How much of the advisor's compensation is attributable to investment analysis, what investment-analysis work was actually performed during the year, and how has that work changed over time?
114. Does the scope of the investment analysis justify the compensation being paid for it?
16. Does the benchmark actually answer the question?
I use benchmarks as information, not as the final answer.
115. What exactly is being compared, and are the plans, services and provider responsibilities actually comparable?
116. Is the benchmark comparing price or value?
117. Does it tell us what the other plans actually receive and how much work their advisors perform?
118. Does it tell us whether their fees are reasonable or whether their compensation structure makes sense?
119. Does it tell us whether their fees are employer-paid or participant-paid?
120. Does it tell us whether their advisors send invoices?
121. Does it tell us whether the services are actually used?
122. Does it tell us what happened to their fees over time?
123. If everyone in the benchmark uses an asset-based fee, does that establish that the fee structure is appropriate?
124. Could the entire benchmark population be using a compensation structure that hasn't been independently examined?
A benchmark is a comparison.
It is not an explanation.
17. Does fiduciary status answer the economic question?
125. Who is the fiduciary, and what are they responsible for?
126. Does fiduciary status tell the sponsor what the service should cost?
127. Does it establish that the compensation corresponds to the work?
128. Does it establish that the fee should be asset-based?
129. Does it establish that participants should pay?
130. Does it establish that an invoice isn't necessary?
131. Does it establish that the advisor's services are necessary or that participants are receiving sufficient value?
132. Does it establish that the provider has negotiated the best price?
133. Does it establish that the employer has considered alternatives?
Fiduciary status matters.
But fiduciary status is not a substitute for examining the economics of the relationship.
18. Does a clean audit answer the question?
134. What was the audit designed to examine, and what wasn't it designed to examine?
135. Did the auditor evaluate whether fees were reasonable?
136. Did the auditor evaluate whether cheaper investment options were available?
137. Did the auditor evaluate conflicts in the investment lineup?
138. Did the auditor evaluate whether the advisory fee corresponded to the work?
139. Did the auditor evaluate whether participant-paid fees were economically appropriate?
140. Did the auditor evaluate the provider relationships?
141. What can we legitimately conclude from a clean audit, and what can't we conclude?
19. What alternatives have actually been considered?
142. Could the plan use a different recordkeeper or TPA?
143. Could it eliminate revenue sharing?
144. Could it eliminate an advisory fee?
145. Could it use a fixed-dollar advisory arrangement?
146. Could the employer pay the fees directly?
147. Could the employer retain the same advisor but change the compensation structure?
148. Could the plan negotiate a lower fee without changing providers?
149. Could it retain the same providers but eliminate unnecessary services?
150. Could another provider perform the same work for less?
151. Could the plan improve without replacing anyone?
If the answer is no, the next question is:
Why are you doing it this way now?
The point isn't to find something wrong
A forensic review shouldn't begin with the assumption that a provider is overpaid, a recordkeeper should be replaced, or a plan needs to be redesigned.
Sometimes the existing arrangement is appropriate.
Sometimes the provider is doing excellent work.
Sometimes the compensation is reasonable.
Sometimes compensation can be negotiated.
Sometimes services should be restructured.
Sometimes a provider should be replaced.
And sometimes the best decision is to do nothing.
The purpose of asking all these questions is not to manufacture a problem.
It is to understand the arrangement well enough to know whether there is a problem.
A benchmark can tell you what other plans pay. A disclosure can tell you what compensation is disclosed. An audit can answer questions within the scope of the audit. A fiduciary designation can establish responsibilities.
But none of those, by themselves, necessarily answers:
What exactly are we buying, what is it costing, who is paying, who is receiving the money, what work are we receiving, how was the arrangement created, and does it still make sense?
That's the investigation I conduct.
The retirement plan is my client.
And before deciding what to change, I want to understand what the plan already has.
A note on where these questions came from
These questions have developed over many years of examining Form 5500 filings, talking with plan sponsors, studying provider compensation, comparing plans over time, and investigating how retirement-plan services are actually purchased. They also build on questions I've written about previously, including how to build a plan from scratch, what services advisors should provide, what benchmarking misses, how fees affect participants, why conventional invoicing matters, and what thousands of plan-sponsor conversations have taught me.
This is not a checklist of questions every plan sponsor must answer. It is a window into how I investigate a retirement plan.