A Hard Rain's A- Gonna Fall (very long)

This past Sunday (April 15, 2018), the New York Times ran an article on its front page (here), right below the fold, that talked about the unsustainability of public employee pensions.  Of course, Oregon PERS was the featured subject of this article, principally because we are one of the few states that publicly disclose the benefits received by all PERS retirees. The title of the article referred to the pension received by Dr. Joseph Robertson, newly retired, former President of OHSU, whose benefit is at the top of the the list.  The article, by Mary Williams Walsh - a long-time Times investigative reporter - was filled with inaccuracies, half-truths, faulty inferences, and a poor history.  The article attracted more than 800 comments, some disputing much of the incorrect information, pointing out errors and omissions, and criticizing the sensationalism of the article.  Of course, there was a near equal number of comments reporting on how bad the public employees were in their own states.  The placement and source of the article (NYT) resulted in it being reprinted, verbatim, in at least 20 different newspapers across the state, and being either reprinted or reprinted with editorials in many of Oregon’s own local newspapers.  I imagine countless other states found reason to reprint the article.  The NYT amplified a poorly written, poorly researched, and frequently inaccurate article at least a thousand fold over what occurs when this same information has been confined to our local “tribal” newspapers.  The effect, I fear, is going to be a much earlier, much more aggressive, and far more antagonistic push by anti-PERS zealots to light everyone’s hair on fire over this.  In turn, the November general election and the subsequent Legislative session will make all of us sense that “a hard rain’s gonna fall."

I want to try, in this post, to correct errors, recharacterize errors of interpretation, and generally try to right the wrongs perpetrated by poor reporting, in what appears to be an Arnold Foundation-funded “hit piece” on public employee pensions in the newspaper “of record” for the United States.  It is sad (to me, anyway) whenever a paper like the New York Times reports on a subject in such a biased way.  I’ve spent my entire adult life depending on the NYT to provide an unbiased record of informed reporting (except, of course, on their editorial page, which is where opinions belong).  When we have an open attack from our current President on “fake news”, it is disheartening to consider the prospect that a major source of “unfake news”  has committed the very sin we’ve been arguing against, and turning a non-story about a couple of outliers into a serious news piece.  Allow me to enumerate:

• The article is entitled "A $76,000 Monthly Pension: Why States and Cities Are Short on Cash”.  It starts with Robertson’s pension benefit and continues its misadventure from there.   First, a significant fact.  There are more than 134,000 PERS retirees and beneficiaries (beneficiaries are not listed in the public list cited in the Oregonian).  A second fact.  Slightly more than 2,000 retirees receive pensions of more than $100,000 per year.  While the second is mentioned; the first is not.  Let’s do the math.  The percentage of retirees earning more than $100,000 comprise roughly 1.5% of ALL PERS retirees.  Put another way, 98.5% of these retirees earn less than $100K (the median is about $27,000 per year, and the average is about $32,000).  Of those who earn more than $100,000, a huge majority are Tier 1 (hired prior to 1996), under rules that have been changed more than twenty times since (although not retroactively, which has been prohibited by the Courts and the IRS).  Moreover, the top 10 recipients (as of 2018) consist of 7 former physician administrators from OHSU, one PSU faculty/administrator who worked for more than 40 years, one Community College President with 38 years of experience, and one successful football coach at the University of Oregon.  The NYT piece focuses on the just-retired President of OHSU and the Football Coach.  The OHSU President worked there for 38 years as a specialist eye surgeon, generating huge billings and income for OHSU, as well holding a series of administrative positions culminating with him being the President of a nationally recognized health sciences complex (Medical School, Dental School, Nursing School, major research center, multiple PhD programs) employing over 20,000, with a multi-billion dollar budget, and generating large sums of income for OHSU and multiplying that through the entire state.  Both he and the football coach were singled out as exemplars of what is wrong with Oregon PERS.  Both had deals (totally permissible under pension guidelines) to count “outside” income in the pension calculations.  In the case of OHSU’s President, his “outside” income came largely from his clinical practice in opthamalogy (eye surgery), which was billed through OHSU at rates including the necessary overhead to pay the PERS contribution on such income.  The football coach received a base salary, plus additional revenue from endorsement deals made with the UO Athletic Department and Nike (which received a $2 billion corporate tax break for keeping its HQ in Oregon).  Again, the endorsement payments were paid to the UO Athletic Department and were (presumably) run through UO payroll, and with the appropriate pension payments made to PERS on all income.  None of this should have come as a surprise to the institutions; nothing was done in secret.  None of this should have surprised the Legislature or PERS, both of which were aware of the details as they were formulated, and as the pension benefits were accruing.  If there was anything untoward, there was plenty of time to object during the negotiations and the signing of the employment contracts.  (And neither recipient is exceptional in salary, being compensated at below average rates for the positions and the institutional “leagues” in which they operate).

• “Oregon’s costs are inflated by the way in which it calculates pension benefits for public employees.  Some of the pensions include income that employees earned on the side [see above].  Other retirees benefit from long-ago stock market rallies that inflated the current value of their payouts.”   The second “problem” closed twice, first in 1996 with the creation of a Tier 2 that does not receive the rate guarantee that Tier 1 members receive.  The second time it closed was at the end of 2003, when ALL employee contributions to Tier 1, Tier 2, and the newly created OPSRP plan were directed to a separate account, not subject to employer match, not subject to any rate guarantee, and subject only to the performance of the market in good times and bad.  The balances in Tier 1 and Tier 2 accounts were frozen as of 12/31/2003 and only earnings have been credited for the past 15 years.  Tier 1 actives and inactives still get rate guarantees on their accounts, but for Active Tier 1 members, salary growth has generally outpaced the earnings on a frozen balance; the vast majority of retirements that have taken place in the past 6 or 7 years have been under Full Formula, not under the “lucrative Money Match” alluded to in the article.  Most of the cases where new retirees earn far more than their Final Average Salaries are instances where the employee has long been inactive, but not retired.  In these cases, the Final Average Salary is frozen at the last known salary when the member worked, while balances continued to grow at the assumed rate until retirement.  Not surprisingly, a large majority of these people end with benefits exceeding their Final Average Salary.  (Note:  the Legislature tried to solve this problem in 2003 by creating a very limited, very short-duration, option to induce inactives to withdraw their money from the system.  The deal was open for slightly more than a year, and offered these members the option to withdraw and rollover their own accounts plus a 50% employer match, forsaking the remaining 50% employer match.  Not many people took advantage of this option because it was poorly advertised, available for a short period of time, and was a bad deal that would have cost members a 25% reduction in benefits accrued toward their retirement).  Nine of the ten top recipients are Full Formula retirees, while one is a Money Match recipient.  So, the article focuses on two Full Formula recipients with unusual (but relatively uncommon) salary arrangements, and blows smoke about Money Match (when only 1 of 10 top recipients receive that benefit).  (The article does acknowledge that Full Formula is what most states use).

• “The bill is borne by taxpayers”.  Of all the claims, this one pisses me off the most.  Of course taxpayers are on the hook for public agency budgets; that’s why they are called “Public Agencies”.  But this overlooks several important points:  1) employers are supposed to submit the amount deducted from employee payroll to pay for employee contributions in real time, and 2) they are supposed to pay their employer contributions at the same time.  If employers had fulfilled their obligations in real time, the current problem would be negligible.  But that didn’t happen.  Following the income tax remedy negotiations, the employers were expected to cover the cost of this in their current contributions.  Moreover, it took the PERS actuary longer than it should have to recognize that Money Match, not Full Formula, had overtaken the formulae calculations starting in the early 1990s.  By 1997, the tax remedy [see below] and the Money Match problems hit the employers simultaneously, with concomitant and hardly unexpected rate hikes to cover the additional expenses.  Instead of just paying them, while the overall economy was growing, the employers caviled, whined, wheedled and cajoled the PERS Board and the Legislature to come up with ways to “amortize” their payments over a longer schedule, “smooth” the ups and downs of the market, and eventually to put rate collars on how much the contributions could grow from biennium to biennium.  In addition, they sued to claw back some earnings crediting to members in 2000 for 1999, setting up the 2003 reforms (see above and below), which, in fact, clawed back 8.67% of the 20% credited earnings from all Tier 1 members who were in the system in April 2000.  In short, the employers, the Board, and the Legislature used creative techniques that allowed PERS employers to deliberately underfund their contributions to the system.  This further overlooks another important factor.  The budgets of many public agencies are built from small taxpayer contributions, coupled with significant user fees levied on the primary beneficiaries of the service.  Higher education, an example I’m most familiar with, started in the 1970s with 90% of revenue coming from the State General Fund (Income Taxes).  By 2017, the state contribution from General Fund dollars (Income taxes) was less than 10% of the operating budgets.  Tuition, grants, contracts, patents, administrative overhead, rentals, fees cover the remaining 90% of the budget - the user fees.  My local city charges me significant water and sewer usage fees, which covers a significant portion of their operating budget.  So, it is a mistake to claim that “…the bill is borne by taxpayers”.  It is partially borne by taxpayers; the remainder is paid by those who use the systems  (think tuition; campground fees; hotel occupancy taxes, water and sewer bills).  My faculty and administrative salaries and benefits were NOT paid from taxpayer money; they were paid from the 90% of the revenue generated from the sources other than tax dollars cited above.  Remember that higher education has been funded with only 10% - 20%  of income tax dollars since the passage of Ballot Measure 5 in 1990.

• The article cites Josephine County as one of the PERS “basket cases”, but fails to mention that Josephine County prides itself on having the lowest property tax rates in the state.  Josephine County refuses to raise its property taxes to support public services, and it is little wonder that they can’t fix roads, jails are at half capacity because of lack of guards, and 911 calls are not promptly acted upon.  One school superintendent complains that PERS is the root of all their problems.  He notes:  “The system (PERS) is run at the state level, but it is bankrolled in large part by obligatory contributions from local governments.”  Again, this superintendent must be completely ignorant of Ballot Measure 5, which passed his county overwhelmingly (and statewide by a smaller margin) in 1990.  Measure 5 did several things (among many others). It limited total property taxes to no more than 1.5% of assessed value (with exceptions for self-inflicted voter-approved levies); Second, it directed that about 90% of school funding come from state general fund revenue (State Income Taxes), while allowing local school districts to assess up to $5 per thousand of assessed value on real property (this is 180 degrees different from pre-Measure 5 rules, where 90% of local school funding came from property taxes and 10% came from the State). Measure 5 also provided no funding mechanism for this massive change in school financing, leaving it to the Legislature to reallocate existing funds to cover the school budgets.  Higher Education and Human Resources were massively underfunded to come up with the revenue to cover K-12 funding.  Worse, the local school boards set the proposed school budgets and negotiated with their public school teachers and support personnel; the district in turn negotiates with the Legislature for their 90% based on the locally established budget.  The state imposes no restrictions on salary and benefit packages for the school districts; they negotiate within the district with the Superintendent and the locally elected School Boards.  As of 2018, all school districts within Josephine County have school tax rates at least 20% below the statutory $5 per $1000 maximum, while the large school districts in the larger areas have maxed out their $5 rate, and have also received approval from the State to form local improvement districts (“local option”) to tax outside the maximum.  Josephine County also has a total effective property tax rate of 0.62% of assessed value (less than half the maximum allowable), while larger counties tax at or above the statutory maximum 1.5% (depending on levies and local options).  The article never once mentions Ballot Measure 5’s hamstringing effect on local funding of schools, much less the paradox of having 90% of the funds coming from the state, yet the state having no control over the local school boards or salary and benefit negotiations of teachers and support personnel in those schools.  Moreover, it never mentions Josephine County’s preoccupation with being anti-tax of any kind.  

• “For decades, PERS calculated pensions two different ways, and retirees could choose whichever produced the bigger numbers.”  FALSE, FALSE, PANTS ON FIRE FALSE.  No member chooses anything about his/her pension except the optional payout method (with a beneficiary, without a beneficiary, lump sum, partial lump sum, annuity certain).  The Oregon statutes clearly require PERS to calculate benefits all applicable ways, and PERS CHOOSES the method yielding the highest Option 1 benefit.   PERS calculates Final Average Salary based on parameters set by the Legislature, and the elements of those parameters that each agency chooses to participate in.  For example, some agencies allow employees to accrue sick leave over their careers as an incentive to reduce absenteeism, and rewards employees by counting “HALF” the value of the sick leave in Final Average Salary calculations; other agencies do not allow its employees to accrue sick leave).  The sick leave calculation is irrelevant to Money Match retirees.  (I left over 3000 hours of accrued sick leave at the table when I retired).

• “…when lawmakers required government retirees to pay Oregon’s 9 percent income tax, as everybody else did, they also increased pensions by 9.89%, giving retirees extra money to pay the tax with.”   Another quarter truth, with a much more complex background, and far less significant effect than dramatized in this article.  First, all Tier 1 members (those hired before 1996), were initially promised (at the time of their hire) that their PERS pensions would NOT be subject to Oregon state income tax.   In 1988, the US Supreme Court decided a case - Davis v Michigan - that concluded that, for income tax purposes, States could NOT treat resident federal retirees differently than public employee retirees.  At the time, Oregon was doing exactly what the Davis v Michigan ruling prohibited.  Thus Oregon was required to either give Federal retirees tax free treatment, or start taxing PERS retirees.  In 1991, the Oregon Legislature started taxing PERS recipients.  They were sued for breach of a statutory contract, impairment of a contract, and wage theft.  The court ruled that the taxation scheme impaired the statutory contract, but that given the State’s decision to tax PERS pensions, then a monetary solution could repair the problem.  After multiple negotiations, lawsuits, settlement agreements, PERS retirees, PERS, the Legislature, Federal retirees, and the Courts reached agreement in 1996 to start reparations beginning in 1997.  At that point, federal retirees were permitted to deduct from their Oregon Income taxes, the amount of pension income attributable to work prior to October 1991 (when the Oregon Supreme Court ruling was finalized), while being taxed on that portion of pension income attributable to work performed after October 1991.  Both Federal retirees and Oregon PERS retirees who retired prior to October 1991, received refunds of taxes paid after the Legislature started taxing PERS pensions.  The PERS retirees whose work was completed prior to October 1991 received an income tax remedy (a pension increase) of 9.89% to cover the state’s then 9% income tax (the 9.89% increase resulted because increasing the benefit by only 9% would have exposed the additional compensation to the very taxation it was supposed to remediate; thus the amount was 9.89% to cover the tax on the 9% remedy).  For those who continued to work after October 1991, their income tax remedy was a fraction of 9.89% representing work performed before October 1991, relative to the length of their entire PERS career.  For me, for example, my work career began in September 1970 and I retired in October 2002.  I received a tax remedy adjustment of 6.6%.  Anyone hired after October 1991 was not eligible for any income tax remedy; their pensions are 100% taxable with no offsetting remedy.  Only Tier 1 members hired before October 1991 are even eligible for any sort of tax remedy.  The article makes this appear as if it were a gift from the Legislature, when, in fact, it was a decision forced on the state by the US Supreme Court, and applicable to only a small subset of current members.

• The article misrepresents Tier 1 and Tier 2 employee accounts.  These are not “tracking” accounts.  They are real accounts whose member contributions were frozen at the end of 2003.  “For workers with the tracking accounts, PERS dialed back the annual returns to 8%, then to 7.5% in 2016”.  WRONG  DEAD WRONG.  First, only Tier 1 was guaranteed any specific earnings - the assumed rate of, then, 8%.  Tier 2 had no such guarantee and has always been credited only with market earnings regardless of the amount.  The maximum annual return ever guaranteed for Tier 1 members was 8%.  Therefore, it wasn’t dialed back TO 8%; it never rose above.  Moreover, the assumed rate (the 8% referenced) had/has a direct bearing on employer contributions.  The higher the assumed rate, the less the employers have to contribute.  So, both the employers and members had a shared common interest in keeping the assumed rate high.  But, beginning in 2014, the rate was lowered to 7.75%, in 2016 to 7.5%, and effective 2018 the rate is now 7.2%.  Each time the rate is lowered, employer contributions have to rise because the system assumes less income from earnings to pay the required pension benefits.  [To be fair here, I *think* the writer intended to say that the 2003 reforms set the assumed rate as the maximum earnings crediting for Tier 1 at 8%, not that it was “dialed back”.  The “dialing back” occurred because the 2003 reforms forbade PERS from crediting regular account earnings above the assumed rate to any Tier 1 member.]

• Finally, the article continues to blame the 2008 stock market meltdown for much of today’s current problem.  While there is no doubt that the 2008 losses exacerbated the problems of PERS and every other pension system in the US, the article omits one salient fact.  By 2007, Oregon PERS was 106% funded; the 2003 reforms began to reduce future costs, capping Tier 1 returns at the guaranteed rate and no more helped considerably, and good stock market returns helped erase the UAL and re-fund the system.  By 2007, PERS had amassed nearly $1 billion in its contingency reserve, an account designed as a rainy day fund to protect the system from unexpected events.  About 2/3 of the way through 2007, the employers were still absorbing the increases that began in 1997, but spread over what seemed like an eternity, and they were still complaining bitterly to the PERS Board about the rate increases.  About this time, the actuaries proposed the concept of rate collaring - a way to keep costs within a more predictable corridor based on the funded status of the system.  After some discussion, the Board adopted the rate collaring method, which limited increases to employer rates to 3% if the system was less than 80% funded; 6% if the funding dropped below 70%.  In addition, the Board (completely reconstituted after the 2003 reforms), agreed to shift most of the contingency reserves (near $1 billion) to employers to further buy down their contribution rates.  (In an exquisitely agonizing piece of irony, the 2000 employer-initiated litigation that led to the 2003 reforms was about the Board FAILING to fund the same reserves adequately).   So, when 2008 rolled around, the stock market tanked by 27%, and PERS and the employers were left with no contingency reserve with which to buffer the 2008 losses.  Of course, the 2008 losses led to further major increases in employer contributions, but they didn’t take effect immediately (as they should have) because of the smoothing (which introduces gains and losses over a prolonged period of time), and the rate collars.  Once again, in 2013, acting against all sensible legal advice, the Legislature, egged on by employers, passed Legislation to reduce the statutory COLA.  This reduction was projected to reduce the UAL by $5.8 billion over 20 years.  In a move that defied all logic, the Legislature not only passed the legislation, but it allocated more than 10% of the projected savings to agencies to spend in the 2013-15 biennium.  Most agencies went on a spending spree with the extra cash instead of using it to further retire their pension obligations.  Just as the 2015 Legislature was ramping up, the Oregon Supreme Court ruled that this, too, was a breach of contract in a unanimous decision, blowing a $580 million hole in the Legislature’s budget, digging the UAL deeper, and paralyzed the Legislature from trying any further efforts to reform PERS.  (As a financial footnote, once 2018 finishes, the 2008 stock performance disappears from typical 10 year averages.  From 2008 to 2017, the overall return for Oregon PERS has been 5.9%, which includes the  -27% return in 2008.  If we assume that 2018 will earn nothing more than the current assumed rate of 7.2%, the 10 year average for the period of 2009-2018 will be about 10.2%, significantly greater than any of the assumed rates during the period).

This is a short (?) rebuttal of many of the points in Walsh’s article.  Sadly, this rebuttal does not have the audience that Walsh had.  Walsh and the NYT managed to magnify the misinformation over an audience more than 1000 times greater than a similar piece written for any local, useless rag.   Moreover, it added credence to the misinformation regularly reported in our local newspapers.  Further, Walsh’s piece has been reprinted in whole or part in at least 20 Oregon community newspapers, with nasty editorials about the “PERS Beast” or the “PERS Dragon” appearing in those and other local outlets.  This has, unfortunately, energized and bolstered the anti-PERS zealots across the state’s media and political chattering class.  As a result, it has made what was likely to be a significant issue in the 2019 Legislative session, into a MAJOR (if not central) issue in political campaigns, including the upcoming May primary, the November election for Governor, for the Oregon House,  one-third of the Oregon Senate, and a Supreme Court seat or two.   So, this leads to the obvious question.  What could the Legislature do?   A lot will depend on the outcome of the November elections and the resulting composition of the Governor’s mansion, the House, and the Senate.  All currently rest in control of the Ds, but control is tenuous.  Neither house has a “supermajority” needed to block or to enact any kind of anti-PERS legislation, or to enact or block any tax revenue measure, including raising the pitifully low Corporate Income tax (did I mention that Oregon has the 6th lowest corporate tax in the US?).  The 2017 Legislature ended in effective gridlock.  All signs point to both parties working exceptionally hard to increase their numbers in the Legislature.  Regardless of how this turns out, I expect the anti-PERS crowd (largely the crew of homies from the Bend/LaPine/Sunriver/Redmond area) to continue their assault on PERS benefits.  I would expect legislation on pension caps (the number $100,000 appears over and over again), forcing an end to the employer “pick up” of the employee contribution, anti-spiking (removing further accumulation of sick leave, vacation time, overtime, comp time, and outside sources of income), and even some legislation intended to either incentivize inactives to cash out of the system, or to force them into accepting smaller pensions.  The worrisome part of these proposals is that the Legislature may just throw up its hands and pass something against rational legal advice and let the Oregon Supreme Court sort it out.  They gambled on this approach in 2013 and lost bigly, digging the hole the system is in today.  The Supreme Court of 2019 will be different from the Supreme Court that ruled unanimously in early 2015 against the retroactive grab at the COLA from the 2013 Legislature.  I was pessimistic going into the 2017 legislature; I’m downright depressed about 2019.   Not only do I expect 2019 to shadow a “hard rain falling”, I also expect “Lawyers, Guns, and Money” because the “shit has hit the fan.”  I wish it weren’t going to get nasty, ugly, and brutish, but I’m expecting the 2019 session to be one of the most contentious in my 48 year history living in Oregon, and PERS will be center-stage in the action.  To make matters even worse, PERS goes in at a real disadvantage this time, as its long-time resident expert and current Executive Director, Steve Rodeman,  retires on June 1, 2018.  The new executive director will come from outside PERS and Oregon, and will go into the 2019 Legislative session at a distinct disadvantage in not fully understanding the system, and all of the complex interactions that any legislative action might produce.  Rodeman excelled at clear communication of anticipated and unanticipated consequences of legislative actions, thus preventing most legal quagmires before they were created.  In addition, the PERS Board loses its Chairman in August 2018.  While the Board Chair does not report to the Legislature, the fact that John Thomas has been one of the most knowledgeable Chairs in PERS history, the new Chair, who will be new to the Board as well, has a steep learning curve in front of him/her and will have to depend more on staff (many also new and inexperienced) for guidance.  This will create an information vacuum that no one will be able to fill quickly enough to do a credible job dealing with the potential fallout from any proposed PERS legislation.  Thus, I expect a lot of anti-PERS legislation to be proposed during this period of extreme political pressure and information/leadership vacuum.  Trust me that revenue reform is the only real long-term solution to Oregon’s problems, but Oregon will only get some sort of a more balanced tax system over the bodies of future PERS retirees.  Worse still, the NYT article may be the wedge that finally creates the schism between the older members with better retirement benefits and the younger employees in the system who have less generous benefits.  I expect the demands for generational equity to create a flash point for legislators and unions representing many public employees in 2019.