Closed DB plan nondiscrimination testing relief extended for 2017


Employers with “closed” DB pension plans (i.e., closed to new entrants, but benefits continue to accrue for existing participants) received another extension of nondiscrimination testing relief in IRS Notice 2016-57. This will help these employers demonstrate compliance with existing nondiscrimination testing rules while waiting for final regulations likely to be effective starting in 2018.


Closed DB pension plans often run into unintended nondiscrimination testing problems. Why? Generally there is higher turnover among non-highly compensated employees (NHCEs) vs. highly compensated employees (HCEs). The NHCEs positions are filled with new employees who are not eligible for the closed DB plan, so the DB plan gradually becomes “concentrated” with HCEs. This will eventually cause the plan to fail the nondiscrimination testing requirement that pension plans not cover a disproportionate share of HCEs.

The IRS acknowledged this unintended consequence with Notice 2014-5 which provided temporary relief for DB plans with a soft-freeze date prior to December 31, 2013. The relief applied to the 2014 and 2015 plan years and allowed aggregated nondiscrimination testing of DC plans and closed DB plans without having to pass the usual “gateway” requirements, as long as the DB plan passed testing on a standalone basis in 2013. This was a significant development because gateway allocations can be very expensive and require per-participant allocations of up to 7.5% of pay. The alternative was to freeze all DB plan accruals so that the coverage requirements no longer applied to the DB plan.

Relief Extended

The relief in Notice 2014-5 was extended to the 2016 plan year in Notice 2015-28, and proposed regulations were issued on January 29, 2016. Since those regulations likely won’t be finalized and effective until the 2018 plan year, the IRS is now extending the temporary nondiscrimination testing relief for the 2017 plan year as well.

Although this relief is welcome news to plan sponsors, they should not lose sight of the fact the closed DB plan still needs to pass nondiscrimination testing when aggregated with the employer’s DC plan(s). This shouldn’t be a problem in most cases, unless the DB plan provides such large benefits to HCEs that the aggregated plans fail the 401(a)(4) benefits testing.

Sponsors of small and mid-sized closed DB plans should also keep in mind that eventually 401(a)(26) minimum participation requirements could become a problem. These rules require that the smaller of (a) 40% of active employees or (b) 50 active employees participate in the DB plan. In high turnover industries, those thresholds may not be far on the horizon.


False alarm! IRS withdraws controversial proposed cross-testing regulation provisions

A couple of months ago, the IRS proposed some changes to the §1.401(a)(4) nondiscrimination testing regulations.  On Thursday, they withdrew the part of those proposed regulations that was bad news for plan sponsors, as noted in our prior post.

We are pleased the IRS has reconsidered the unintended consequences benefit formula restrictions and “facts and circumstances” based determinations could have on employers’ willingness to sponsor qualified retirement plans.

Surprises in the proposed cross-testing regulations

surprised face

On January 29th, the IRS proposed revisions to the nondiscrimination testing regulations of §1.401(a)(4).  The title of proposed regulations (and most of the attention generated by them) is focused on the relief for closed defined benefit (DB) plans.  This post summarizes the proposed changes that would affect more than just closed DB plans.  Most of them are beneficial to plan sponsors, but one is not.

The bad news – benefit formula restrictions

The biggest surprise is a proposed restriction in setting different benefit levels for different participant groups.

Currently, plans can generally separate the participant population into groups with different benefit levels/formulas as desired, so long the plan is doesn’t disproportionately favor Highly Compensation Employees (HCEs) relative to non-HCES.  Plans can even go so far as to separate each plan participant into his or her own “group”.

The proposed regulations would require HCEs’ benefit formulas to apply to a “Reasonable Classification” of employees*.  This is a “facts and circumstances” determination.  §1.410(b)-4(b) states that “reasonable classifications generally include specified job categories, nature of compensation (i.e. salaried or hourly), geographic location, and similar business criteria”.  Picking participants by name (or in a way that effectively does that) is not considered a reasonable classification.

This is an important issue for plans that allow each participant to have a separate benefit level and rely of the Average Benefit Test to satisfy §1.401(a)(4).  Other plans may need to consider if their benefit groups are a reasonable classification.

*Unless the rate group satisfies the Ratio Percentage Test.

The good news – cross-testing gateways for aggregated DB/DC plans

The favorable part of the proposed regulations is more flexibility in combining DB and DC plans for nondiscrimination testing.  These proposed changes were suggested by the IRS in Notice 2014-5, so they aren’t a surprise to those that have kept up with the IRS’s previous relief efforts for closed DB plans.  However, those suggestions haven’t got much attention so they are good news for many.

Plans must pass through a “gateway” before aggregating DC and DB plans in a cross-test.  A cross-test is generally much more favorable than testing each plan separately.  The currently available gateways are:

  1. The DB/DC plan is “primarily defined benefit in character”
  2. The DB/DC plans consist of broadly available separate plans
  3. The DB/DC plan provides a minimum allocation to all benefitting non-HCEs

The proposed regulations would expand the DB/DC gateway options in three ways:

1.  New gateway: The proposed regulations would add another gateway – passing the cross-test test with a 6% interest rate (rather than the standard 7½% to 8½%). While the DB/DC would technically still need to pass the cross-test with a standard interest rate, this option could practically eliminate the gateway requirement for DB/DC plans that can pass with 6% interest.

2.  Matching contributions use: The proposed regulations would allow the average matching contribution for non-HCEs (up to 3% of pay) to count toward the DB/DC minimum allocation gateway. Matching contributions would still not be included in the cross-test.

3.  Option to average DC allocation rates: Current rules allow DB allocation rates for non-HCEs to be averaged for satisfying the minimum allocation gateway. The proposed regulations would allow the same treatment for DC allocation rates.  The purpose of this change is to allow plans to provide lower allocation rates for those with less service by providing higher rates to those with more service.

The IRS notes that they’re considering if restrictions on this option are needed to ensure it is used as intended, and not as another technique for minimizing non-HCE benefits.  The proposed regulations would also limit averaging of DB and DC rates to reduce the impact of outliers.

Many plans that satisfy the §1.401(a)(4) requirements with a general test will need or want to revisit their benefit formula design if these proposed regulations become final.  There is sure to be a lot of resistance to the benefit formula restrictions, so the regulations may not be finalized as proposed.  If you would like to send comments on the proposed regulations you can do so until April 28, 2016.

Plan sponsors may apply the proposed regulations specific to closed DB plans right away, but may not use the flexibility of the other proposed cross-testing rules until they are finalized.  We encourage you to contact your actuary if you have questions about how these proposed rules would affect your plan.

What’s the Impact of 2016 IRS Retirement Plan Limits?

The IRS just announced the 2016 retirement plan benefit limits, and there are virtually no changes from 2015. What does it all mean for employer-sponsored retirement plans? Below is a table summarizing the primary benefit limits, followed by our analysis of the practical effects for both defined contribution (DC) and defined benefit (DB) plans.

Qualified Plan Limit 2015 2016
415 maximum DC plan annual addition $53,000 $53,000
Maximum 401(k) annual deferral $18,000 $18,000
Maximum 50+ catch-up contribution $6,000 $6,000
415 maximum DB “dollar” limit $210,000 $210,000
Highly compensated employee (HCE) threshold $120,000 $120,000
401(a)(17) compensation limit $265,000 $265,000
Social Security Taxable Wage Base $118,500 $118,500


Changes affecting both DB and DC plans

  • Qualified compensation limit remains at $265,000. A flat qualified compensation limit could have several consequences. These include:
    • More compensation counted towards SERP excess benefits if a participant’s total compensation (above the threshold) increases in 2016.
    • Lower-than-expected qualified pension plan accruals for participants whose pay is capped at the 401(a)(17) limit and were hoping for an increase.
  • HCE compensation threshold remains at $120,000. For calendar year plans, this will first affect 2017 HCE designations because $120,000 will be the threshold for the 2016 “lookback” year. When the HCE compensation threshold doesn’t increase to keep pace with employee salary increases, employers may find that more of their employees become classified as HCEs. This could have two direct outcomes:
    • Plans may see marginally worse nondiscrimination testing results (including ADP results) if more employees with large deferrals or benefits become HCEs. It could make a big difference for plans that were previously close to failing the tests.
    • More HCEs means that there are more participants who must receive 401(k) deferral refunds if the plan fails the ADP test.

DC-specific increases and their significance

  • The annual DC 415 limit remains at $53,000 and the 401(k) deferral limit remains at $18,000. Although neither of these limits has increased to allow higher contributions, it should make administering the plan a little easier in 2016 since there are no adjustments to communicate or deal with. Since the 401(k) deferral limit counts towards the total DC limit, this means that an individual could potentially get up to $35,000 from profit sharing ($53K – $18K) if they maximize their DC plan deductions.
  • 401(k) “catch-up” limit remains at $6,000. Participants age 50 or older still get a $6,000 catch-up opportunity in the 401(k) plan, which means they can effectively get a maximum DC deduction of $59,000 ($53K + $6K).

DB-specific increases and their significance

  • DB 415 maximum benefit limit (the “dollar” limit) remains at $210,000. This limit remained unchanged for a third straight year, which may constrain individuals with very large DB benefits (e.g., shareholders in a professional firm cash balance plan) who were looking forward to increasing their DB plan contributions/deductions.
  • Social Security Taxable Wage Base remains at $118,500. When this limit increases, it can have the effect of reducing benefit accruals for highly-paid participants in integrated pension plans that provide higher accrual rates above the wage base. When the wage based remains unchanged (like this year), it means that these individuals’ accruals may be higher-than-expected if their total compensation (above the wage base) continues to increase.

OPEB Funding Policy Opportunities

New Governmental Accounting Standards Board (GASB) statements 74 and 75 are intended to make accounting for OPEB (Other Postemployment Benefits, usually retiree medical) more transparent by moving the entire unfunded liability to the face of the financial statements. This post discusses some of the OPEB funding policy opportunities that employers should consider as they prepare to implement GASB 74/75 over the next two years.

Opportunity #1: The potential for funded status improvement. Unlike pension plans which are generally pre-funded, most public sector OPEB liabilities are unfunded and benefits are pay-as-you-go. But employers should consider that ANY funding policy is better than nothing – and even a modest level of pre-funding will improve the plan’s funded status and balance sheet impact.

Opportunity #2: A funding policy can help lower the calculated liability. The ultimate cost of retiree medical benefits can never be known exactly until benefits are actually paid in the future. However, the best estimate of those costs in today’s dollars is impacted by how (and if) funds are invested. To the extent that investment returns can help fund future benefit costs, an OPEB trust with higher expected returns can help reduce your calculated OPEB liability.

Opportunity #3: Attention can drive action. OPEB liabilities are becoming front-page headlines – particularly because of their unfunded nature. Employers should harness this increased scrutiny to compel stakeholders (e.g., employers, employees, and taxpayers) to collaboratively review OPEB terms and create a strategy to prudently prefund these promises.

There will be pitfalls along the way. It may be difficult to secure additional funding sources when employers and employees are already trying to eliminate existing pension debt. Plus, OPEB costs are usually more volatile than pensions, so an OPEB funding policy will need to consider strategies to deal with this “moving target”.

There may also be situations where an employer does not want to prefund OPEB. Perhaps they fear that “committing” money towards OPEB will reduce their ability to reduce benefit levels in the future. However, this rationale just skirts the larger issue of knowingly promising benefit levels which are unaffordable.

GASB 74/75 is a catalyst for public employers to revisit their OPEB funding policies. There is a limited timeframe for stakeholders to develop a meaningful funding strategy before the entire unfunded liability goes “on the books”.

Developing a thoughtful public pension funding policy

A thoughtful pension funding policy provides the best way for public sector employers to keep their pension promises. The funding policy provides discipline to address unfunded pension liabilities, while respecting the many current demands for budget dollars.

The League of Arizona Cities and Towns recently sponsored a study by a special Task Force to identify areas for improvement and develop reform recommendations. The final recommendation is “Create a pension funding policy.” The Task Force emphasized that Arizona cities and counties “have a fiduciary responsibility to ensure (the) plan has sufficient financial resources to provide the benefits earned ….”

The pension funding policy is the best (and maybe the only) way for plan sponsors to meet their fiduciary duty to keep the pension promise. But developing a pension funding policy is not easy. It requires projecting unfunded pension liabilities beyond what is usually shown in existing actuarial reports. It also requires a delicate balancing of all of the various constituent interests over a variety of time frames.

The easy way out has always been to ignore the problem and leave it to someone else in the future. But a prudent few will openly address the political and budget issues inherent in developing a funding policy that keeps the pension promise.

One city recently chose to invest the time and resources to develop a pension funding policy. After years of declining funded ratios and increasing contributions, the city Finance Officer championed a study that involved the City Commission, the pension fund members, the actuary and the investment advisor. The resulting pension funding policy provided for an increase in employer and employee payroll contributions – and also provided for a third stream of contributions unrelated to payroll: a pension sustainability contribution.

We encourage every public sector plan sponsor to develop a pension funding policy. And we applaud the prudent finance officers who actually begin the process to keep their pension promises.

First pensions, now OPEB – New GASB 74 & 75 will transform OPEB reporting

Public sector employers, get ready! The Governmental Accounting Standards Board (GASB) has officially approved new accounting statements for Other Post-Employment Benefit plans (OPEB; retiree medical). Here’s what the recent GASB announcement confirms:

  • Final provisions will closely mirror GASB 67/68 pension accounting. The official statements won’t be released until late June, but last summer’s OPEB exposure draft included similar provisions such as:
    • The unfunded liability will now go on the balance sheet.
    • The liability discount rate will be based on a projection of how long dedicated assets (plus future contributions) will cover current plan members’ future benefit payments.
    • OPEB expense will recognize asset and liability changes over a shorter time period.
    • Goodbye ARC! Funding and accounting are officially separated, so plans and employers should consider developing a new OPEB funding policy.
  • Expanded note disclosures and RSI are required, including sensitivity of results to a +/-1% change in the discount rate and medical trend assumptions.
  • Effective dates:
    • GASB 74 Plan accounting is first effective for reporting periods beginning after June 15, 2016 (e.g., fiscal years beginning July 1, 2016 or January 1, 2017).
    • GASB 75 Employer accounting is first effective for reporting periods beginning after June 15, 2017 (e.g., fiscal years beginning July 1, 2017 or January 1, 2018).

Although the implementation dates are almost 3 years away, employers should take action now to prepare. Some questions to ask include:

  1. How can I develop and implement an OPEB funding policy over the next few years? Prefunding OPEB can help reduce the unfunded balance sheet liability.
  2. What plan benefit adjustments and investment policy changes are available to lower my long-term OPEB liability? Now is the time to review your OPEB management strategies.
  3. What is my strategy to educate stakeholders about OPEB promises and their potential financial impact under the new GASB 74/75 requirements? OPEB is a complex topic that may be unfamiliar to plan members, government decision-makers, and taxpayers.

The new GASB 74/75 statements should help make OPEB promises more understandable and transparent. While there’s still sufficient lead-up time, employers should view this as an opportunity to proactively address an employee benefit which is often unfunded but must be managed prudently.

Pension Lump Sums Much More Expensive in 2015

Over the past few years, many defined benefit (DB) plan sponsors considered lump sum payouts to their terminated vested participants as a way of “right-sizing” their plan. The ultimate goal is to reduce plan costs and risk.

The IRS recently released the November 2014 417(e) rates, which will be the 2015 reference rates for many DB plans. This post shares a brief update of the impact these rates could have on 2015 lump sum payout strategies.

Low Interest Rates Will Increase Cost of 2015 Lump Sums

So, what’s the potential impact on 2015 lump sums? The table and chart below show the possible difference between comparable 2014 and 2015  lump sums at sample ages assuming payment of a $1,000 deferred-to-65 monthly benefit. Note that the 2014 lump sum estimates are based on November 2013 interest rates, while 2015 values are based on November 2014 rates.

November 2014 lump sum chartNovember 2014 lump sum tableNote: If we adjust for the fact that participants will be one year older in 2015 (and thus one fewer years of discounting), then this increases the costs above by roughly another 5% at most ages.

What’s Causing Lump Sum Costs to Increase?

DB plans generally must pay lump sum benefits using the larger of two plan factors:

(1) The plan’s actuarial equivalence; or

(2) The 417(e) minimum lump sum rates.

Since interest rates have been so low over the past few years, the 417(e) rates are usually the lump sum basis. In particular, 2013 lump sums were abnormally expensive due to historically low interest rates at the end of 2012 (the reference rates for 2013 lump sum calculations), while higher rates towards the end of 2013 made 2014 lump sums more affordable. This is because lump sum values increase as interest rates decrease, and vice versa.

For calendar year plans, the lookback month for the 417(e) rates is often a couple of months before the start of the plan year. Here’s a brief comparison of the November 2013 rates (for 2014 payouts) versus the November 2014 rates (for 2015 payouts).November 2014 segment rate tableAs we can see, the first segment rate increased slightly while the second and third segment rates decreased substantially since last November. The overall large decrease in interest rates is why lump sums will be more expensive in 2015.

What Else Should Plan Sponsors Consider?

  1. If you’re still considering a lump sum payout window, you’ll want to carefully weigh the additional costs of the 2015 lump sum rates compared to 2014.
  1. Even with lower interest rates pushing up lump sum costs, there are still incentives to “right-size” a plan now. These include (a) large ongoing PBGC premium increases and (b) the potential for new mortality tables to further increase lump sum costs significantly in a couple of years.
  1. In addition to lump sum payout programs, plan sponsors should consider annuity purchases and additional plan funding as ways to reduce long-term plan costs/risks

DB Plan Sponsors Should Prepare Now for Higher Year-End Liabilities

The combination of lower discount rates and new mortality tables will dramatically increase pension plan liabilities and decrease DB plans’ funded status for December 31, 2014 financial reporting. Using the November 2014 Citigroup Pension Liability Index (CPLI) and Citigroup Pension Discount Curve (CPDC) as proxies, pension accounting discount rates are down by almost 90 basis points since December 31, 2013.

Fortunately, many plans have experienced solid investment returns so far during 2014. This will take some of the sting out of the liability increases, but it likely won’t be enough to entirely offset the effect of lower interest rates and the new mortality tables. The higher liabilities will affect both the year-end funded status of the plan and also the 2015 pension expense calculation.

Discount Rate Analysis

In the chart below we compare the CPDC at three different measurement dates (12/31/2012, 12/31/2013, and 11/30/2014). We also highlight the CPLI at each measurement date. The CPLI can be thought of as the average discount rate the CDPC produces for an “average” pension plan.

Citigroup comparison 11302014

The orange arrows in the chart highlight the trend in yield curve movement and show how rates are almost back to their 2012 lows at all points along the spectrum. This means that nearly all plans will feel the negative effect of lower discount rates.

Net Effect on Balance Sheet Liability

Depending on the starting funded status, the change in pension liabilities and assets can have a leveraging effect on the reported net balance sheet asset/liability.

Below is a simplified illustration for a plan that was 80% funded on 12/31/2013, where we assume a 10% increase in pension liability during 2014. We then compare the funded status results under two asset scenarios: (1) Assets 5% higher than 12/31/2013 and (2) Assets 8% higher than 12/31/2013.

11302014 bal sheet liability example

In both cases, the funded status of the plan decreases. There’s also a magnified increase in the unfunded balance sheet liability because it’s such a leveraged result. This amount increases by 30% and 18%, respectively, in the two sample scenarios.


So, what should plan sponsors be considering over the next month as we approach year-end? Here are a few ideas.

  • Don’t forget that the new Society of Actuaries mortality tables will be recommended for use at year-end and will likely further increase plan liabilities.
  • Additional pension plan funding (above the IRS minimum requirements) may be appealing in 2014 and 2015. Not only will it increase the plan’s funded status, but it will also help lower your pension plan’s PBGC variable rate premiums.
  • Your plan’s specific cash flows could have an enormous impact on how much the drop in discount rates affects your pension liability. If you’ve just used the CPLI in the past, it’s worth looking at modeling your own projected cash flows with the CPDC or an alternative index or yield curve to see how it stacks up.
  • Now may be a good time to consider strategies that lock in some of this year’s investment gains. These could include exploring an LDI strategy to more closely align plan assets and liabilities, or offering a lump sum payout window for terminated vested participants early in 2015.


What’s the Impact of 2015 IRS Retirement Plan Limits?

The IRS just announced the 2015 retirement plan benefit limits and we’re seeing some modest increases from 2014. What does it all mean for employer-sponsored retirement plans? This post analyzes the practical effects for both defined contribution (DC) and defined benefit (DB) plans, followed by a table summarizing the limit changes.

Changes affecting both DB and DC plans

  • Qualified compensation limit increases from $260,000 to $265,000. Highly-paid participants will now have more of their compensation “counted” towards qualified plan benefits and less towards non-qualified plans. This helps for both nondiscrimination testing as well as for benefits.
  • HCE compensation threshold increases from $115,000 to $120,000. For calendar year plans, this will first affect 2016 HCE designations because $120,000 will be the threshold for the 2015 “lookback” year. Slightly fewer participants will meet the new HCE compensation criteria, which will have two direct outcomes:
    • Plans may see better nondiscrimination testing results (including ADP results) if there are fewer participants at the low end of the HCE range, especially those with big deferrals. It could make a big difference for plans that were close to failing the tests.
    • Fewer HCEs means that there are fewer participants who must receive 401(k) deferral refunds if the plan fails the ADP test.

DC-specific increases and their significance

  • The annual DC 415 limit increases from $52,000 to $53,000 and the 401(k) deferral limit increases from $17,500 to $18,000. A $1,000 increase to the overall DC limit and $500 increase to the deferral limit may not seem like much, but it will allow participants to get a little more “bang” out of their DC plan. This means that individuals can get up to $35,000 from employer match and profit sharing ($53K – $18K) if they maximize their 401(k) deferrals. Previously, their profit sharing limit would have been $34,500 ($52K – $17.5K).