What New Jersey Actuaries Say Public Pensions Cost

Andrew Biggs had a piece in Forbes where he made the point:

Overall, state and local governments dedicated just 4.1% of their spending to pension contributions in 2013. The reality, though, is that if state and local governments were truly fully funding their pensions – that is, merely following the rules that the federal government requires for corporate pensions – their annual contributions would make up over 20% of total state and local government spending.

But contributions made by governments are not a good measure of what is being accrued in benefits since:

  1. a major part of the Annual Required Contribution (ARC) for most governments (New Jersey being in the forefront) consists of paying off unfunded liabilities generally over 30 years (over and over again),
  2. member contributions offset the ARC, and
  3. many governments (New Jersey being in the forefront) do not, being broke, pay the full ARC.

All of which makes projecting the true cost of these benefits using reasonable assumptions problematic.  One would need to go to the actuarial reports and pull out the contribution exhibits – which is what I did with the New Jersey reports – and come up with a spreadsheet that yields this data:

  1. 442,827 active participants with an average salary of $60,738
  2. $3 billion annual cost ($6,775 per participant) which is offset by $2 billion that the employees put in
  3. $5.2 billion being the amortization payment means that a 7.9% interest rate and a 30 year repayment period puts the unfunded liability at $60 billion

and these observations:

  1. $6,775 for a comparatively generous defined benefit plan is ridiculous.  That’s IRA contribution territory.  The real cost is in the $15,000 range per active participant which translates into a total cost for annual accruals of about $7 billion.
  2. Not making the ARC (plus accrued interest) over the years ($15 billion and counting during the Christie interregnum alone) would explain how part of that $60 billion underfunding developed but what about the rest of it? Could it be bad actuarial assumptions?
  3. It is and it’s continuing.

 

 

17 responses to this post.

  1. Posted by Tough Love on April 6, 2016 at 4:11 am

    I believe you $15K avg. contribution should be closer to $17.5K even w/o COLA reinstatement

    Reply

  2. Posted by Tough Love on April 6, 2016 at 11:09 pm

    John,

    Given that THIS Blog-article is about ….”What Pensions Cost” …. I thought it would be instructive to expand on the Normal Costs shown in your linked spreadsheet….. by way of one example, your Column “Police/Fire Local”:

    Your spreadsheet shows a Total Normal Cost of 19.32% of salary, split (via taking ratios from #s in your spreadsheet) EE=9.37% and ER=9.95%.

    In response to a question I asked you on your prior post, you responded that these spreadsheets are based on NJ’s valuation assumptions, including the 7.9% interest rate.

    I created a spreadsheet based on a Local PFRS Officer retiring after 25 years of service and then living either 25 or 30 years. These 2 scenarios should encompass the life expectancy of most Officers retiring after 25 years within the usual range of retirement ages.

    The following spreadsheet inputs and assumptions were used:

    (1) first year salary = $25,000
    (2) 25 year salary = $100,000
    (3) the rate of salary increase through year 6 was twice that of the following years. I did this to acknowledge the much more rapid wage increases in the first 5 to 7 years.
    (4) Employee contributions are 9.5% of pay (rounding from your 9.37% EE normal cost in 2015) in all years.
    (5) Where I show results with COLA-increased pension payouts, the assumed COLA increase is 2% in each year.
    (6) the Officer’s starting pension (after 25 years of service) is 65% of final pay
    (7) the investment earnings rate is the SAME in both the accumulation and payout periods, and 3 different rates were tested:
    (a) the Average rate used by Private Sector pension Plans … 3.8%
    (b) the rate that Moody’s deems appropriate and uses in it’s analysis … 5.5%
    (c) NJ’s “official” valuation” assumption …. 7.9%

    The following percentages represent the Employer’s (i.e., the Taxpayer’s) level annual %-of-pay “Normal Cost” (in addition to the 9.5%-of-pay Employee contribution) to fully fund the 65% of pay pension over the Officer’s 25 working years, and than exhaust the accumulated fund by the end of the 25 (or 30) year payout period.
    ——————————————————–
    Taxpayer Contribution – 25 Year Payout

    3.8% without COLA …………… 38.7%
    3.8% with COLA ……………….. 50.0%

    5.5% without COLA …………… 24.1%
    5.5% with COLA ……………….. 31.2%

    7.9% without COLA …………… 10.8%
    7.9% with COLA ……………….. 14.6%
    ———————————————————–
    Taxpayer Contribution – 30 Year Payout

    3.8% without COLA …………… 44.0%
    3.8% with COLA ……………….. 59.0%

    5.5% without COLA …………… 26.9%
    5.5% with COLA ……………….. 36.0%

    7.9% without COLA …………… 11.9%
    7.9% with COLA ……………….. 16.6%
    ————————————————————-

    Interestingly, for my 2 scenarios using 7.9% without COLA (which is consistent with the development of your spreadsheet figures) MY resultant Employer Normal Costs of 10.8% (with 25 year payout) and 11.9% (with the 30 year payout) are only slightly higher than the 9.95% Employer Normal cost in YOUR spreadsheet. I would guess that my %s are slightly higher because, although there is a VERY low “quit rate” among NJ Police, some Officers serve for fewer than 25 years and get a pension with a somewhat lower “Normal Cost”. This occurs due to the back-loaded nature of typical DB pension Plans.

    The closeness of MY spreadsheet Normal Cost results under the scenario …..7.9% without COLA …… to YOUR spreadsheet Normal Cost figure, validates my “process”, thereby providing a mechanism to determining what the Normal Costs would be under all 12 Scenario combinations shown above.

    I expect that the VERY significant INCREASE in Normal Cost as the interest rate assumption DECREASES will surprise many, but THAT underlies the very vocal discussions of the need for material pension reform.

    And of Note…… only in the LOWEST Normal-cost scenario (7.9% without COLA) where the Taxpayer Normal Cost is 10.8% or 11.9% (assuming 25 or 30 year payout periods respectively) are the Taxpayer’s contribution to these Officer’s Plans even remotely close to what Private Sector workers typically get in retirement contributions from their employers (3% to 4% into a 401K Plan, plus their employer’s 6.2%-of-pay into Social Security on their behalf …. totaling about 10%-of-pay).

    In all of the OTHER scenarios, Taxpayers must contribute 2, 3, 4, 5, even 6 times MORE than what they get from their employers.

    This grossly excessive, unnecessary, and unfair pension structure of HEADS the Public Sector workers “win” and TAILS, the Taxpayers “lose” rightfully must end …. and for the future service of all CURRENT workers.

    Reply

  3. Posted by S Moderation Douglas on April 7, 2016 at 7:08 am

    Them there is real scary numbers. Biggs is even scarier:

    “I found that, using corporate pension rules for valuing liabilities and amortizing unfunded liabilities, the average state and local pension contribution would rise from 24% of employee wages to 105%. That’s how big and expensive these plans are.”

    105% of wages.

    ONE HUNDRED AND FIVE PERCENT OF WAGES!!!

    And he’s not just talking safety pensions, he’s talking …all… pensions combined.

    105% of wages. “That’s how big and expensive these plans are.”

    Except, they really aren’t, on an ongoing basis.

    Although he is rather ambiguous, he is not talking about a “normal cost” of 105 percent. He is talking normal cost …plus… amortization of unfunded liabilities …within seven years.

    Even scarier, he’s talking “average” state and local. Fresno CA, being fully funded under “government” rules, might need, using corporate pension rules, to increase contributions to 35 percent, from the present 15 percent. (That’s pure spit ball example.) And after a few years, once the pension is fully funded using 5.5 percent discounting, their normal cost might revert back to 15 percent.

    New Jersey state systems, being one of the worse funded systems, would have to contribute MUCH more than the “average” 105% of wages. Not to pick on New Jersey specifically. Illinois, Pennsylvania, Connecticut, and, yes, California and others, under Biggs scenario, would all obviously be “above average” and hypothetically pay more than 105 percent.*

    For a few years, then ARCs would return to a more normal level. (Higher than they now are, but NOT 50% or 59%, as in your “math”. Not even for cops. Not even in New Jersey, or California. Not even close. )

    *”Hypothetically”, because it would never happen. It’s just an example, guys.

    Reply

    • Posted by Tough Love on April 7, 2016 at 1:15 pm

      SMD,

      Biggs 105% of wages INCLUDES BOTH the “Normal Cost” and the amortization of the current unfunded liabilities over 7 years (the period used in Private Sector Plans).

      Once the payoff of unfunded liabilities is paid off, year 8 and later costs would be ONLY the Normal Cost (of year 8 accruals) as shown in my above comment, plus some actuarial gains & loses from the year just ended. And if using the 3.8% rate Biggs deems correct, the those Normal Costs would be 50% (w/o COLAs) and 59% (with COLAs).

      The 105% *as well as the very high Normal Cost (many multiples greater than what Private Sector workers get from their employers) is the “price to pay” for VASTLY over-promising and being unable to pay for those promises.

      A SCARY number is not a WRONG # ….. but certainly an item that YOU feel needs to addressed via your classic “smoothing”. …… nothing to see here folks, we’re not ripping off the Taxpayers too much.

      Reply

  4. Posted by Tough Love on April 7, 2016 at 1:29 pm

    SMD,

    Quoting ……

    “In an interview with the Financial Analysts Journal last year, Nobel laureate economist William F. Sharpe, creator of the Sharpe ratio for risk-adjusted investment performance analysis, said public pensions in the United States are a “disaster” and “a crisis of epic proportions.””

    http://www.ocregister.com/articles/pension-710925-government-pensions.html
    ——————-

    Looks like it’s time for some more ……. “smoothing”.

    Reply

  5. Posted by Smooth Moderation Equal on April 7, 2016 at 2:06 pm

    “Biggs 105% of wages INCLUDES BOTH the “Normal Cost” and the amortization of the current unfunded liabilities over 7 years (the period used in Private Sector Plans).”

    Duhh!!!

    That’s exactly what I said…

    “He is talking normal cost …plus… amortization of unfunded liabilities …within seven years.”
    ———————————————
    “And if using the 3.8% rate Biggs deems correct, the those Normal Costs would be 50% (w/o COLAs) and 59% (with COLAs).”

    I

    Don’t

    Think

    So,

    Tim

    Biggs and others …may…* have a point in using a risk free rate to calculate the future “value” of a DB pension, but I don’t think he or anyone else advocates using that to calculate the normal cost for purposes of funding. Unless one actually chooses to “invest” in only risk free investments.

    Ask Mr. Biggs if his own 401(k), IRA, etc. is entirely in risk free investments.

    1) Past Performance Is No Guarantee of Future Results.

    2) We are in very unusual financial and political times right now.

    But… I think most experts would agree that if we contributed 59% of wages (plus 9 point something from the employee). AND invested in a reasonable asset allocation, a fully funded system would probably soon be over funded. At that point, even ERISA would not require the full normal cost based on 3.8%. Or even 5.5%, depending.

    Absent the second coming, or it’s financial equivalent, no DB system nowhere no how will ever require a constant 59% of wages over time.

    *may? That will be argued elsewhere.

    Reply

    • Posted by Tough Love on April 7, 2016 at 2:22 pm

      Quoting SMD ……

      “I think most experts would agree that if we contributed 59% of wages (plus 9 point something from the employee). AND invested in a reasonable asset allocation, a fully funded system would probably soon be over funded. ”

      I don’t disagree (although over the LONG term), but what you choose to ignore in making such statements is the value of the “guarantee”. Public Sector workers want the BENEFIT of the high return (such as NJ’s 7.9%), but don’t want to (and now do not) assume any risk that that return does not materialize ….. via passing on the bill for any “shortfall” in the return (from that 7.9%) to the Taxpayers.

      Effectively, NJ’s Public Sector workers get a GUARANTEED 7.9% return*. Where in this (or most other) investment environments can anyone (OTHER THAN Public Sector workers) gets such guarantees ?

      * and it’s even WORSE, because Public Sector workers get a GUARANTEED 7.9% return on the value of their Pension LIABILITIES …… of which today, about 50% of which have ZERO assets supporting those liabilities.. Where do you get a 7.9% returnn WITHOUT investing ANYTHING ?

      It’s doesn’t work that way outside the fairyland of Public Sector pensions.

      —————————————-
      Quoting SMD …. “I don’t think so”.

      I really doubt that a person who’s career responsibilities included changing light bulbs is “qualified” to make such determinations.

      Reply

  6. Posted by Smooth Moderation Equal on April 7, 2016 at 5:46 pm

    I dint ignore nothin.

    “Moderation, fair and balanced.”

    My mission (among others) is to point out inaccurate and/or misleading statements. Mr. Biggs quote “….. the average state and local pension contribution would rise from 24% of employee wages to 105%. That’s how big and expensive these plans are.”

    Is exactly the kind of statement that gets misconstrued on Brietbart and Fox news or Unionwatch, or the like. Yes, if these plans actually required an ongoing 105% of wages, they would be “big and expensive” and yes, I intentionally left off the first part of the first sentence (” I found that, using corporate pension rules for valuing liabilities and amortizing unfunded liabilities,…..”) because half the people reading that don’t know what it means, and because half the reporters or pundits quoting Biggs either don’t understand it, or would intentionally leave it out.

    And 59% of wages as an ongoing normal cost is incorrect, also.

    We know that you never get tired of making incorrect statements, but don’t you ever get just a little bit tired or ashamed of being needlessly rude?

    On pension blogs my name has been “Moderation” for several years now. In real life or social networks I’m considered “Mellow”. At one time or another, I’ve been insulted by some of the best and some of the worst. Mellow doesn’t mind. It rolls off like water off a duck’s back. Mellow considers the source. Sometimes it’s funny; sometimes just sad. Either way, it’s more of a reflection of you than of me, FWIW.

    Reply

    • Posted by Tough Love on April 7, 2016 at 7:21 pm

      SMD,

      You likely were good at “balancing” (on ladders) while changing all those light bulbs….. but when it comes to the justification and need for PUBLIC Sector Pension Reform (yes, meaning material pension REDUCTIONS for all CURRENT workers), your opinions are anything but “balanced”.

      It’s all about ……………..”smoothing”……………. isn’t it ?

      In THIS case you focus on that 105% contribution requirement, which Mr. Biggs CLEARLY identifies as being calculated using the 3,8% interest rate, AND including BOTH the Normal Cost as well as amortizing the current unfunded liability over 7 years …… noting that these are the IDENTICAL to the standards REQUIRED (by the US Government) of Private Sector Pension Plan valuations.

      It seems clear that by repeating (in multiple comments) these SAME things, your INTENTION is to distract the readers from the HUGE “Normal COST” when NOT using the very high (7.9% in NJ) “official” valuation interest rates, but instead, using rates (Biggs 3.8% or Moody’s 5.5%) that everybody EXCEPT public Sector Unions/workers/retirees (with a vested interest in the status quo) believes to be appropriate.

      And repeating from my above comment, THESE are the NJ Local Police/Fire Taxpayer-share of JUST the Normal Cost:

      ========================================
      The following percentages represent the Employer’s (i.e., the Taxpayer’s) level annual %-of-pay “Normal Cost” (in addition to the 9.5%-of-pay Employee contribution) to fully fund the 65% of pay pension over the Officer’s 25 working years, and than exhaust the accumulated fund by the end of the 25 (or 30) year payout period.

      Taxpayer Contribution – 25 Year Payout

      3.8% without COLA …………… 38.7%
      3.8% with COLA ……………….. 50.0%

      5.5% without COLA …………… 24.1%
      5.5% with COLA ……………….. 31.2%

      7.9% without COLA …………… 10.8%
      7.9% with COLA ……………….. 14.6%
      ———————————————————–
      Taxpayer Contribution – 30 Year Payout

      3.8% without COLA …………… 44.0%
      3.8% with COLA ……………….. 59.0%

      5.5% without COLA …………… 26.9%
      5.5% with COLA ……………….. 36.0%

      7.9% without COLA …………… 11.9%
      7.9% with COLA ……………….. 16.6%
      ————————————————————-

      Reply

  7. Posted by Smooth Moderation Equal on April 7, 2016 at 10:47 pm

    You’re repeating yourself.

    Reply

    • Posted by Tough Love on April 8, 2016 at 12:02 am

      It’s necessary to counteract your endless attempts to move the ficus and discussion AWAY FROM demonstrations, facts, and yes MATH ……. that CLEARLY supports the need for very material pension reform.in NJ.

      Reply

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: