SB0722: SUMMARY OF SUBSTITUTE BILL IN COMMITTEE (Date Completed: 12-2-14)

Senate Bill 723

The bill would allow for an entity other than ORS to administer Tier 2 plans and for an entity other than ORS to contract with personnel necessary to implement Tier 2 plans. (This change would be necessary since Senate Bill 722 would allow local employers to designate their own providers of Tier 2 plans, other than the provider chosen by ORS.)

The bill is tie-barred to Senate Bill 727.

Senate Bill 724 (S-1)

For all service credit purchases other than military service, the bill would eliminate the option to purchase service credit beginning on the effective date of the bill. Specifically, the bill would eliminate the purchase of out-of-system public education service credit (Sec. 69); the purchase of service credit for service performed at a nonpublic school (Sec. 69c); the purchase of up to five years' general service credit (Sec. 69f); the purchase of service credit for State of Michigan service (Sec. 70); and, the purchase of service credit for sabbatical leave, employee organization professional services leave, or employee organization professional services released time (Sec. 71). (As noted below, under current law, individuals hired after July 1, 2010, may not purchase service credit.)

For the purchase of service credit for military service, the bill would require the amount charged for the purchase to be based on actuarial costs instead of the current requirement to pay 5% of salary for each year purchased (Sec. 74).

Senate Bill 725

The bill would eliminate disability retirement allowances and duty disability retirement allowances for any member hired on or after July 1, 2014.

Senate Bill 726 (S-1)

The bill would provide for 90% health care (medical, dental, and vision) premium coverage for eligible retirants, beneficiaries, and dependents with household incomes below 133% of the Federal poverty level, instead of the current 80% coverage (unless a retiree was Medicare-eligible as of January 1, 2013, in which case coverage is 90%). An eligible retirant includes a person who first became a member of the system before July 1, 2008; after that date, a graduated premium is in effect for employees hired between July 1, 2008, and September 3, 2012, after which all new hires are part of the "401k for health care" savings plan, rather than premium coverage.

The bill further would require the Department (i.e., ORS) to develop a method for the retirant, beneficiary, or dependent to prove income eligibility, and would direct ORS to use income tax returns as the basis for determining eligibility. The bill also would require ORS to use reasonable controls consistent with the State's income tax return audit policies to ensure that eligibility was maintained. The bill further would require ORS to develop a method for identifying individuals eligible for the 90% premium coverage by examining an individual's income tax return in the year before the year of retirement and to automatically apply the 90% premium if feasible. By December 31, 2015, ORS would have to submit its proposed plan to implement these changes to the Legislature.

Senate Bill 727 (S-1)

The bill includes three major changes. The first relates to the closure of the existing hybrid (defined benefit and defined contribution, or DB and DC) retirement plan and its replacement with a defined contribution-only plan. The second would lower the rate cap that local units pay on the unfunded accrued liability (UAL) by one percentage point each year until the cap was lowered from the existing 20.96% of payroll to 9.96% of payroll. Finally, the bill would eliminate any possibility of the payment of a "13^th check". Each of the three major changes is discussed below.

Closure of Existing Hybrid; New DC-Only Plan

The bill would eliminate the existing hybrid (DB and DC) retirement plan and would place all people hired on or after the effective date of the bill into a defined contribution-only plan. These members would immediately vest in their own contributions and would vest in the contributions made to their 401k plans by their employer on the following schedule: after two years, 50%; after three years, 75%; and after four years, 100%. However, members would have to annually meet with an investment provider selected by ORS in order to vest in employer contributions.

The DC plan offered under this version of the bill would provide for a 66% employer match for employee contributions up to 6% of pay (or, a maximum employer match of 4%) for members with fewer than 20 years of service. After 20 years of service, the bill would provide for a 100% employer match for employee contributions up to 7% of pay. An employee would be allowed to contribute more than 6% or 7%, but the employer would be capped at 4% matching contributions if an employee had fewer than 20 years of service, and 7% if an employee had 20 or more service years.

Analysis of Employee with Full Retirement

Table 1 provides a look at an individual's hypothetical savings comparing the existing hybrid and the proposed DC-only plan. The calculations assume the employee has a starting salary of $30,000 that grows 3.5% per year, starts work at age 25, works 35 years, and retires at age 60. The calculations assume 7% returns on Tier 2 (i.e., 401k) contributions; assume monthly pre-tax retirement spending of $5,280 (which was chosen to exhaust the 401k savings accumulated under the hybrid at age 85); and assume straight life pension benefits (which means no survivor pension benefits). The analysis does not discount for net present value, but is a simple compounding in today's dollars.

The analysis calculates the value of the hybrid's pension plus DC assets (recalling that the DC portion of the hybrid provides a 1% employer match on 2% employee contributions), and compares that to the value of the individual's DC assets, assuming that the employee invests the same amount in a DC-only plan as the individual contributes to the hybrid plan for both the DB benefit and the DC contributions. The total a person contributes to the hybrid is roughly 7%-8% of salary, which is 5%-6% to pay for the pension component of the hybrid and 2% contributions into the DC component to generate the 1% employer match.

It is important to note that in the above scenario, if a retiree died earlier than the projected time frame, and had chosen a straight-life payout, the pension benefit would cease. Any remaining 401k contributions under the hybrid plan would be passed on to survivors, as would any 401k contributions under the proposed DC-only plan. However, for each plan participant, savings level, spending level, and age of death, the comparison of assets remaining to pass along to survivors would vary, although presumably for the first several years after retirement, there would be larger 401k savings remaining under the proposed DC-only plan than under the hybrid.

Conversely, it is important to note that a DC-only plan provides benefits based on contributions into the plan and relies on investment returns on those contributions. In the above scenario, a 7% return on investment was assumed, since this is the assumed level of investment return for the assets in the DB component of the hybrid plan. However, any deviation from the 7% assumption directly affects an individual's DC portfolio, while the employer absorbs and pays for any shortfall in earnings on the DB component of the hybrid.

Table 1

Existing Hybrid (Straight Life) vs. Matching DC-Only Plan after 35 Years
	Existing Hybrid	Proposed DC-Only Plan
Annual Pension Payment	$47,412/year	$0
Total DC Account Balance at Time of Retirement	$188,820	$778,470
Break-Even Point	Assuming monthly pre-tax spending in retirement of $5,280, the hybrid employee would be able to afford that spending level by combining pension and his or her DC account balance, which would be exhausted after 25 years (age 85), assuming an annual 7% return on investments. However, if the employee remained alive after age 85, the DB component of the hybrid would continue to provide a monthly benefit of $3,950. For the DC-only employee, monthly spending of $5,280 would be affordable and the employee would have $168,238 remaining on hand after 25 years (age 85), under an assumed 7% annual return on investments. The $5,280 monthly spending level would last until age 90. A monthly spending level of $5,280 represents 65.6% of pre-retirement income, under the scenario analyzed. The break-even" point is roughly 90 years of age, where the DC plan would provide a greater benefit if the retiree died before that time and the DB plan would provide a greater benefit if living past that time.

Source: SFA Calculations

If a hybrid employee chose one of several survivor benefit options, then the pension benefit would continue on after the retiree's death, at the chosen benefit level, although pension payments would be smaller in the years that the retiree was alive. Again, age at time of death, benefit level chosen, savings, and spending levels all will vary and make easy comparisons difficult to summarize.

If a hybrid employee chose 100% survivor benefits (which means that whatever the retiree receives during retirement will continue on after death) in the above scenario, monthly pension benefits would be reduced approximately 15%. However, this is an actuarial calculation so the remaining 85% of benefit payments during the retiree's lifetime and then paid during the survivor's lifetime would be projected to yield the same level of total benefits.

Using the above scenario, if one assumes that a hybrid retiree is a married male with a life expectancy of age 82 who chooses 100% survivor benefits, then if his wife lives to age 85, she would continue to receive benefits after his death until her own. Specifically, looking at the actuarial tables for the Michigan Public School Employees Retirement System (MPSERS), it appears that the pension benefits would be roughly 85% of what otherwise would be paid under 100% straight life, which would reduce pension benefits early on, but, again, they would continue after the retiree's death. Because these are actuarially equivalent, there would be no difference in determining a break-even point. The DC-only plan would provide more income in the early years after retirement, but the hybrid plan would continue paying a pension after the retiree died. Actual experience of any given retiree and his or her spouse or dependent would affect the outcomes.

Analyses of Employee with Less than Full Retirement

The first scenario analyzed an employee who starts in the hybrid plan and works for 35 years. However, there are differences if the analysis is broadened to review employees who leave with fewer years of service, particularly those who leave before vesting in any type of pension plan (i.e., fewer than 10 years of service). In the proposed DC-only plan, once an employee had four years of service, he or she would be 100% vested in both the employee's own contributions and the employer match. However, under the hybrid plan, if an employee leaves before 10 years of service, he or she do not receive any pension benefit, and only receives his or her contributions into the pension component of the hybrid plus any accumulated savings in the 401k component of the hybrid.

Table 2 below analyzes the available savings for an employee who leaves after five years of service (i.e., does not vest in the pension component of the hybrid), and Table 3 analyzes the available savings and pension for an employee who leaves after 15 years of service. The assumptions for both scenarios regarding age, wages, and contributions are the same as discussed earlier for the full retirement scenario described in Table 1.

Table 2

Existing Hybrid vs. Matching DC-Only Plan after Five Years
	Existing Hybrid		Proposed DC-Only Plan
Annual Pension Payment	$0		$0
Total DC Account Balance after First Five Years assuming 7% return	$5,525 DC balance plus $8,201 in refunded contributions (assuming 1% interest), for a total of $13,725 DC balance		$20,249 DC balance
Total DC Account Balance at age 60 (assuming no additional contributions past the first 5 years and 7% return on balance yearly)	$104,489	$154,143

Table 3

Existing Hybrid (Straight Life) vs. Matching DC-Only Plan after 15 Years
	Existing Hybrid	Proposed DC-Only Plan
Annual Pension Payment	$10,212/year	$0
Total DC Account Balance after first 15 years assuming 7% return	$27,866 DC balance	$103,872 DC balance
Total DC Account Balance at age 60 (assuming no additional contributions past the first 15 years and 7% return on balance yearly)	$107,833 DC balance	$401,952 DC balance

Lowering the Employer Rate Cap on Unfunded Accrued Liabilities (UAL)

Under current law, employers pay the entire normal cost of a retirement plan, but are capped at 20.96% of payroll for the portion of retirement costs related to unfunded accrued liabilities. Any amount above 20.96% required to be remitted to fund the system each year is paid from the School Aid Fund (SAF). For FY 2014-15, that cost above the 20.96% rate cap sums to $731.0 million appropriated from the SAF in the K-12, Community Colleges, and Higher Education budgets.

The bill would roll back the 20.96% rate cap by one percentage point each year until the cap was 9.96% of payroll, which would take 11 years. The amount resulting from the reduction in the rate cap and corresponding costs to local employers would be paid for off the top of the School Aid Fund, at roughly $100.0 million per year per percentage point, which would accumulate over time, such that the additional cost 11 years out would be roughly $1.1 billion yearly until the UALs were paid off (anticipated to be 23 years from now).

Eliminating the "13^th Check"

The third change in the bill would eliminate the possibility of paying out any "13^th checks". Under current statute and practice, if the retirement system's assets are sufficient to fully sustain all promised benefits, and investment returns in a given year exceed 8%, then an additional one-time pension payment may be made to retirants or beneficiaries. This bill would eliminate this possibility.

Senate Bill 728

The bill would make the investment of an individual's accumulated employer and employee contributions and earnings to one or more investment choices within available categories of investment provided by the Department subject to the provisions in Section 131a. Senate Bill 722 would amend Section 131a to allow a reporting unit to adopt an alternative retirement plan, other than those provided by the Department, and Senate Bill 728 would allow the investment of dollars in those alternative plans.

The bill is tie-barred to Senate Bill 727.

MCL 38.1431a (S.B. 722)

38.1425 (S.B. 723)

38.1369 et al. (S.B. 724)

38.1386 & 1387 (S.B. 725)

38.1391 (S.B. 726)

38.1305 et al. (S.B. 727)

38.1427 (S.B. 728)

FISCAL IMPACT

Senate Bills 722 and 723: Locally Chosen Tier 2 Plans

These bills would increase costs to the Office of Retirement Systems by requiring ORS to hire a third party administrator of all locally chosen Tier 2 plans. In addition, local employers could see additional administrative or commission costs associated with choosing local contractors for the provision of Tier 2 products compared with using the Tier 2 administrator currently providing Tier 2 services, due to a loss of economies of scale and purchasing power that are otherwise available under larger or statewide contracts.

Senate Bill 724 (S-1): Eliminating Purchase of Service Credit

For anyone hired after July 1, 2010, the purchase of service is prohibited. Any changes made in service purchases via statute would affect only the older population in the closed plans, which will be a declining potential pool year after year. Also, with the exception of Sections 69, 71, and 74, all other allowable service credit purchases are required to be actuarially sound. Therefore, there would be little fiscal impact associated with their elimination.

Sections 69, 71, and 74 (out-of-system, sabbatical service credit, and military) require a member to remit an amount equal to 5% multiplied by his or her previous year's compensation multiplied by the number of years purchased. The actuarial cost for these types of service credit purchases is similar to the actuarial costs for universal service credit purchases, and range based on age and years of service. However, a rough average looks to be about 20%; therefore, the per-contract savings from this bill would be 15% on average, applied to a person's wages, multiplied by the number of service years purchased. The average out-of-system contract cost is $9,126, so 15% multiplied by that amount would yield savings of $1,369 per contract. The average sabbatical service contract cost is $2,088, and 15% of that figure would yield savings of $313 per contract. The average military contract cost is $8,583, and 15% of that figure would yield savings of $1,287 per contract.

There are not many service contracts purchased under Sections 69, 71, and 74; the most recent data available showed nine sabbatical purchases, 215 out-of-system purchases, and 38 military purchases in 2014. Therefore, the total fiscal impact of eliminating them, while resulting in savings per contract as estimated above, would be fairly small, and would be roughly $360,000 in the first year. This would be a declining savings pool because new hires in the hybrid system are not allowed to purchase service credit.

Senate Bill 725: Elimination of Disability

The most recent Comprehensive Annual Financial Report (CAFR) indicates that there were 6,072 people receiving disability benefits and the benefits paid totaled about $65.4 million in 2013. The bill would eliminate disability benefits for any newly hired employees. Therefore, over time, as existing members stopped drawing disability payments, the cost would decline because no new eligible members could qualify for disability. There would not be immediate savings of $65.4 million, because anyone hired before the date specified in the bill still could qualify for disability. The bill, however, would not allow any new employees to receive disability benefits upon a determination of becoming disabled.

Senate Bill 726 (S-1): 90% Health, Dental, Vision Care Coverage for Income <133% FPL

Because data are unavailable, estimating the cost of this legislation is difficult. While it would be possible to determine how many retirees' pensions are less than 133% of the Federal Poverty Level (FPL), presumably there are MPSERS retirees whose total household income is greater than just their MPSERS pension.

Extrapolating from 2011 income tax returns and the number of people claiming the senior exemption that year, it is possible that the same percentage of people claiming the exemption and with income less than 133% of the FPL would be a reasonable proxy for the percentage of MPSERS retirees who would have total income of less than 133% of the FPL. In this case, roughly 40% of all single and joint filers had income in 2011 (the last year of the senior exemption) that was less than 133% of the FPL. If the same percentage were to hold true in the MPSERS retiree pool, then 40% of retirees and their beneficiaries would qualify for 90% coverage instead of 80%.

According to the 2013 CAFR, members contributed $394.8 million toward retiree health, dental, and vision benefits, which represents the amounts retirees are required to pay under current law. (Non-Medicare and Medicare-eligible but retired after 2012 pay 20%; Medicare-eligible and retired before 2013 pay 10%.) At this time, there are roughly 67,336 non-Medicare retirants and beneficiaries paying 20% and 20,075 Medicare retirants (who retired after 2012) and beneficiaries also paying 20%, and this is the pool that would be affected by the change to 10% payment if income eligibility were met. (The other 154,840 retirants and beneficiaries receiving health care subsidies under the system were Medicare-eligible as of July 1, 2013, and already pay not more than 10%.)

If 40% of the 67,336 non-Medicare group and 40% of the 20,075 post-2012 Medicare group had to pay 10% instead of 20%, then the additional cost to employers would be estimated at $35.4 million. Table 4 below shows the new payment that an income-eligible retirant or beneficiary would pay under this bill, using 2014 posted insurance rates.

Table 4

Change in Health Care Payments if Income Eligible for 90% Coverage
Category	Number	Current Payment	New Payment if Income Eligible
Non-Medicare
Self	35,730	$1,654.56	$827.28
Spouse	22,876	$3,046.08	$1,523.04
Children	8,730	$3,656.04	$1,828.02
Medicare Post-2012
Self	12,500	$602.88	$301.44
Spouse	7,569	$1,158.72	$579.36
Children	6	$1,756.80	$878.40

Source for Category and Number: ORS; posted insurance rates used for Current Payment

Senate Bill 727 (S-1):

Graduated DC Plan to Replace Hybrid DB/DC

Assuming all new employees would fully contribute the 6% DC contribution during their first 20 years of service, and then the full 7% after 20 years, and receive 4% in matching employer contributions for the first 20 years and 7% matching thereafter, the additional normal cost once the entire system became part of the DC-only plan is estimated at 1.04% of payroll. This is derived by looking at the distribution of workers today and how many have fewer than 20 years of service (who would receive the maximum 4% match) and how many have 20 or more years of service (who would receive the maximum 7% match).

On a distributional basis, about 78% of current employees have fewer than 20 years and 22% have more than 20 years. When applying those ratios to the DC match rates, this would mean average contributions of 4.71% over the lifetime of any given employee under the plan prescribed by this legislation. The cost to employers of the hybrid plan today is the sum of the pension normal cost (2.67%) plus the 1% hybrid DC cost, for a total hybrid cost of 3.67%. Therefore, the difference between the average contributions of 4.71% under the proposed DC-only plan once fully implemented and the 3.67% for the existing hybrid plan is 1.04% of payroll, which would be an additional cost to employers.

The additional normal cost of 1.04% yearly would not immediately be felt, however, because new employees would not have 20 years of experience, and would generate at most a 4% employer cost, which is only 0.33% above the hybrid cost. However, long term, when the system would be fully incorporated over several generations, the 1.04% cost differential would be present, assuming the distributional basis of today's workforce and years of service were fairly constant over time, and participation was 100%, and could approach the $120 million to $130 million/year range, assuming payroll growth over time.

If one looks at a cost differential of 0.33% for the next 20 years (the difference between the 4% DC for all employees because they would all have fewer than 20 years of service compared to the hybrid normal cost of 3.67%) and then a cost differential of 3.33% for 22% of the population as they earn a 7% match for more than 20 years of service, and a 0.33% differential for the other 78% of the population for the next 10 years, the additional cost to employers is estimated at $1.2 billion over the next 30 years.

If participation were not 100% in the full matching plan, these estimates would be adjusted by the actual participation. The State Employees Retirement System' (SERS) DC plan is currently assuming 100% contribution on the 3% match when determining appropriations for budgets. However, if the actual participation in MPSERS were something less than 100%, then there would be a difference in costs.

For example, 80% participation would mean a long-term 3.77% DC normal cost (80% of the 4.71% weighted average) compared to 3.47% total hybrid cost (which is 2.67% normal hybrid cost plus 80% of the 1% DC-match), or a difference of 0.30%. When the system was fully implemented, this 0.30% additional normal cost could approach $37 million to $40 million/year, assuming payroll growth over time.

However, in the short-term, because 80% of a 4% DC match is 3.20% compared to the hybrid's 3.47% cost with 80% participation, an 80% participation rate would result in savings compared to the existing hybrid, until the system was fully mature. In the immediate term, an 80% participation rate in the proposed graduated matching DC plan would result in estimated first-year savings of $1.0 million, but when the 7% match began after 20 years, there would be a slight amount of additional normal costs, and the estimated first 30-year costs overall would be $74.0 million. Table 5 summarizes these potential fiscal impacts.

Table 5

Summarizing Fiscal Impacts of S.B. 727 (S-1), Various Assumptions: Normal Costs Shown are as Compared to Existing Hybrid
	4% matching for up to 20 years of service; 7% thereafter
First-Year Impact
100% participation	$1.3 million cost, or 0.33% of payroll
80% participation	$1.0 million savings, or 0.27% of payroll

First 30 Years Impact (as entire system shifts to DC)
100% participation	$1.2 billion cost, or 0.33% of payroll for first 20 years and then a blended average for next 10 (78% of population at 4% match; 22% of population at 7%)
80% participation	$74.0 million cost, or 0.60%

Fully-Mature System	Assumes Weighted Average Cost of 4.71% (78% of entire population at 4% match; 22% of population at 7% match)
100% participation	1.04% of payroll additional normal cost of DC plan
80% participation	0.30% of payroll additional normal cost of DC plan

It should be noted that the dollar estimates provided in Table 5 and the text above rely upon estimates made for payroll in the system over the next 30 years and beyond. To the extent that the actual payroll deviates from the estimates, the dollar impacts shown above also would fluctuate. The percentages, however, would not change.

[Please see the PDF version of this analysis, if available, to view this image.]

In addition to the normal costs outlined in Table 5, if the hybrid plan were closed to new hires, the Office of Retirement Services has indicated that, even though the Governmental Accounting Standards Board does not "require" accelerated payments, it is still an actuarially recommended best practice to accelerate funding for a closed system for cash flow reasons, among other reasons. According to ORS, the actuary has indicated that the additional unfunded accrued liability contributions over the next four years would be greater than $2.0 billion if the hybrid were closed to new hires.

From the employer's perspective, moving to a DC-only plan would eliminate the potential for future unfunded accrued liabilities that could occur in a defined benefit plan if market performance were less than the assumed 7% rate of return or if other actuarial experience deviated from actuarial assumptions, but could cost more on a yearly normal basis depending on plan structure and participation rates. From an employee's perspective, a DC-only plan can be more portable, but risk is assumed entirely by the employee.

According to a 2012 analysis from the Office of Retirement Services, if the rate of return in the hybrid system were lowered from 7% to 6%, it would raise the normal cost by 1.5 percentage points. At the present time under the 7% assumed rate of return, there are no unfunded accrued liabilities in the hybrid system, but if the assumed rate of return were lowered, then the hybrid's normal costs would increase. According to the 2013 valuation, the rate of return in the hybrid since its inception in 2010 was 8.11%; the rate of return in the closed, legacy system averaged 4.38% over the last 10 years. Between 1970 and 2010, the average rate of return was 9.23% for the assets in the MPSERS portfolio.

Lowering UAL Rate Cap

As mentioned earlier, the cost to the State via the School Aid Fund of lowering the rate cap from 20.96% by one percentage point each year would be roughly $100.0 million per year per percentage point, on top of the existing rate cap costs already appropriated and provider for in statute. Table 6 below shows the cumulative and long-term additional estimated costs from this proposal, assuming that the existing 24-year amortization period is maintained. While it would cost more to the State, local employers would see an equal reduction in costs associated with this proposal, until the UAL was paid off.

Table 6

Additional Estimated Costs to the School Aid Fund of the Rate Cap Proposal
Fiscal Year	UAL Rate Cap	Cumulative Cost to SAF
2015-16	19.96%	$100.0 million
2016-17	18.96%	$200.0 million
2017-18	17.96%	$300.0 million
2018-19	16.96%	$400.0 million
2019-20	15.96%	$500.0 million
2020-21	14.96%	$600.0 million
2021-22	13.96%	$700.0 million
2022-23	12.96%	$800.0 million
2023-24	11.96%	$900.0 million
2024-25	10.96%	$1.0 billion
2025-26 through 2038-39	9.96%	$1.1 billion
2039-40 and beyond	0.00%	$0

Source: Senate Fiscal Agency calculations assume that 1% applied to payroll of $10.0 billion average over time is a $100.0 million cost per percentage point per year. Variations would occur with actual payroll, actual UAL, and amortization over time as experience is compared to assumptions.

Elimination of "13^th Check"

The elimination of paying out a "13^th check" should have little fiscal impact on the system, since the practice occurs only if and when the system is fully funded and investments exceed an 8% return.

Fiscal Analyst: Kathryn Summers

This analysis was prepared by nonpartisan Senate staff for use by the Senate in its deliberations and does not constitute an official statement of legislative intent.