Featured Post

Was it constitutional for Proposition 124 to replace PSPRS' permanent benefit increases with a capped 2% COLA?

In this blog I and multiple commenters have broached the subject of the suspect constitutionality of PSPRS' replacement of the old perma...

Monday, August 29, 2016

The waiting is the hardest part for PSPRS members: The only update the the Hall case is that there is no update

I know that the number one question that everyone has (myself included) is when will the Hall v. EORP decision come out.  This is repeatedly asked in the comments and on the Facebook page.  If you are a Tier 1 member (20 years of service before January 1, 2012) or a Tier 2a member (hired before January 1, 2012 but with less than 20 years as of that date), there are major financial implications for you.  A decision favorable to the plaintiffs would mean refunds in the thousands of dollars to Tier 1 and Tier 2a PSPRS members and a 4% decrease in contribution rates going forward.  The flip side is that PSPRS will go further into deficit and affect the budgets of all public safety employers.  A decision in favor of the defendants will maintain the status quo but will also set a precedent that contribution rates are not fixed at the time of hire.  This could mean theoretically that the Arizona Legislature could constitutionally raise employee contribution rates as high as they want.  Remember that the recently enacted pension reforms require Tier 3 members (those hired July 1, 2017 or after) to split contributions 50/50 with employers, as is done in the Arizona State Retirement System.  This 50/50 split could not realistically be imposed on Tier 1 and Tier 2a/2b members as it would leave employees with little or no net pay, but it could certainly be possible for the employee contribution rate to creep up a few percentage points.  The Hall decision may affect some retirees, especially if you were an active PSPRS member after SB 1609 was enacted and did not DROP, but I would expect that this would be a very small number of individuals.  (Note: The Parker v. PSPRS case is contesting the same issue as Hall, and the decision in Hall will decide Parker as well.)

I wish I had some insight into when the decision was coming.  If anyone, other than the four judges and the one Arizona Supreme Court justice who heard the case in February 2016, knows when or what the decision will be, I don't know who that would be.  My understanding of the process is that once the decision is finalized and ready to be made public a 24-hour notice is given to the parties to the case.  That is it.  I do not know if there will be also be a 24-hour advance press release or any other public announcement, though such a big decision will be headline news the day of the decision.

For some perspective, we can look at the timeline of the Fields case.  The Arizona Supreme Court heard arguments in Fields on June 4, 2013 and didn't announce a decision until February 20, 2014, which is over 8 months. The arguments before the panel in Hall were on February 18, 2016, a little over 6 months ago, so if they take as long as in Fields, we could be waiting until October 2016 for a decision. Furthermore, Hall is dealing with the issue of increased contribution rates (the COLA issue in the Hall case is now moot with the passage of Proposition 124 in May 2016), an issue that is not as clearly protected by the Arizona Constitution, so if they take longer than in Fields and announce a decision sometime in 2017, I would not be surprised, especially in light of the aforementioned consequences to all parties.

I have added a link to the Arizona Supreme Court in the sidebar.  If you go to active civil cases, you will see Hall v. EORP at the very top of the list.  It shows the last update to the case was in April 2016 with nothing else since that date.  I have heard that PSPRS is preparing for a plaintiff victory, though I do not know if this is just wise preparation or if they have some other insight into what will happen.  Ideally, a plaintiff victory will come soon enough before the end of 2016 that refunds would come in the 2016 calendar year.  I say "ideally" because my guess would be that the large refunds of excess contributions will likely be subject to  an IRS withholding rate of 25%.  (I do not know how the state of Arizona will treat the refund; hopefully more leniently since it was the state's doing in the first place.)  These refunded contributions are wages and subject to state and federal income tax, and the lump sum nature of the payment is similar to a bonus.  A payment in 2016 will allow members to get back any overpayment in taxes, if they have a lower taxable rate than 25%, early in 2017 via their 2016 tax refund.  A payment in early 2017 would mean members would not see a refund of any tax overpayment until a year later after they file their 2017 taxes.  It is way too early to speculate about the refund, but I would guess that it would come back through the employer if the PSPRS member is still employed by a public safety agency.  Also, the refund could also be broken up into smaller chunks or paid out over two or more years to lower the immediate tax burden.  Once again, we are probably getting a little ahead of ourselves here.

A decision could come tomorrow or six months from now.  I wish I had more information, but we will all just have to wait.

Thursday, August 25, 2016

PSPRS investment returns through June 2016, or how PSPRS made more for Wall Street than it did for PSPRS members and Arizona taxpayers

The following table shows PSPRS' investment returns, gross of fees, versus the Russell 3000 through June 2016, the final month of the current fiscal year (FY), with the FY end 2014 and 2015 returns included for comparison:

Report PSPRS PSPRS Russell 3000 Russell 3000
Date Month End Fiscal YTD Month End Fiscal YTD
6/30/2014 0.78% 13.82% 2.51% 25.22%
6/30/2015 -0.73% 4.21% -1.67% 7.29%

7/31/2015 0.13% 0.13% 1.67% 1.67%
8/30/2015 -1.43% -1.31% -6.04% -4.47%
9/30/2015 -1.02% -2.31% -2.91% -7.25%
10/31/2015* 1.95% -0.36% 7.33% 0.08%
11/30/2015 0.37% 0.09% 0.55% 0.63%
12/31/2015 -0.95% -0.86% -2.05% -1.43%
1/31/2016 -1.41% -2.26% -5.42% -7.00%
2/29/2016* -0.35% -2.61% -0.52% -7.52%
3/31/2016 3.45% 0.84% 7.04% -0.48%
4/30/2016 0.64% 1.49% 0.62% 0.14%
5/31/2016 -0.11% 1.39% 1.79% 1.93%
6/30/2016 -0.32% 1.06% 0.21% 2.14%

PSPRS did not provide monthly investment returns for October 2015 or February 2016, and returns for those months are estimates based on the prior and following months' returns. There is usually about a two-month lag in PSPRS reporting its investment returns.

PSPRS has reverted back to form for the FY end, ending with a final FY return, gross of fees, of 1.06% versus 2.14% for the Russell 3000.  The last post about PSPRS returns covered the returns through April 2016, and at that time, PSPRS was actually showing a higher return than the Russell 3000.  PSPRS did not have a July 2016 Board of Trustees meeting, so the May and June 2016 returns were included in the August 2016 meeting materials.  Since April PSPRS has lost about a third of a percent while the Russell 3000 gained two percent.

PSPRS earned 49.53% of the Russell 3000 in FY 2016.  For FY 2014 and FY 2014, PSPRS earned 54.80% and 57.75% of the Russell 3000, respectively.  For the three and five year periods, PSPRS earned 55.97% and 50.86% of the Russell 3000, respectively.  PSPRS has shown a pattern of earning just 50-60% of the Russell 3000 since PSPRS implemented its "nationally recognized" investment strategy, though it looks like PSPRS is working its way toward only a 50-55% range.  With a 7.40% expected rate of return (ERR), the Russell 3000 would have to average between 13.50% and 15% annual returns in order for PSPRS to earn that 7.40% ERR.

If we look at individual asset classes, only the private equity class earned at or above the ERR with FY 2016 earnings of 12.41%.  The next highest earner was private credit at 4.23%.  Of the ten major categories, seven had positive returns while three had negative returns.  Seven of the ten categories did not achieve their target benchmarks.  PSPRS missed the fund target benchmark of 1.98% by 0.92%.  What I found most interesting was the real estate category.  At the end of April, it had a healthy FY year-to-date return of 7.95%, but it ended the FY with a return of only 1.69%.  Real estate lost 1.87% in May and a whopping 4.00% in June.  Much of PSPRS returns for FY 2016 were driven by just two categories, private equity and real estate--two asset classes that are difficult to value since they are not easily liquidated.  It leaves us wondering how real estate dropped so much over such a short time.  Barring a major economic crisis that would affect all asset classes, we would expect real estate to be very stable and not subject to such monthly volatility.

The cherry on top of all this is that these FY 2016 returns are gross of fees.  PSPRS usually pays about a half-percent in aggregate investment fees, which means that, net of fees, PSPRS will have earned more for its investment "advisors" than it did for PSPRS members and Arizona taxpayers.  A lesson from Randolph and Mortimer Duke might be in order here for PSPRS' investment staff.  In anticipation of the excuses that will soon be coming for PSPRS Board of Trustees Chairman Brian Tobin, Administrator Jared Smout, and Chief Investment Officer Ryan Parham about why PSPRS had such bad returns, the next post will go little more into PSPRS' returns.

 * Returns, gross of fees, are used because PSPRS usually does not report returns, net of fees paid to outside agencies, except on the final report of the fiscal year.  Returns, gross of fees, are used in the table for consistency.  The past two years fees have reduced the final annual reported return by about a half percent.  Returns, net of fees, were 13.28% and 3.68% for fiscal years 2014 and 2015, respectively.

Monday, August 15, 2016

PSPRS service purchase calculations and how they can cost you a bundle if you're not careful

PSPRS has posted the new service purchase estimator that will be used until the start of next fiscal year.  You will need Microsoft Excel to open it on your own computer.  I would encourage anyone who can buy service time to play around with the calculator just to see the dollar amounts required to buy their service time, now or in the future when it becomes much more expensive.

The calculator has some nice features.  It allows you to break down fractional time into a yearly total, so if you have, for instance, four years, nine months, and 28 days, it will convert that into 4.827 years for you.  I have not looked at my DD-214 in years, but if I remember correctly, it breaks down military service into years, months, and days, which now can be easily converted for your calculation.

There is also a calculator for those who may be interested in purchasing service time via payroll deduction.  Once you calculate the total service purchase cost, you can enter that total into another calculator that will give you a biweekly payment amount.  This biweekly amount can be paid for and length of time up to 15 years for a total of up to 390 payroll deductions.  The payroll deduction also gives you a breakdown of how much you will pay in principal, interest, and in total.

The one feature I do not like about the calculator is how they fix dates for the purchase of time on either the current date or July 1, 2017.  I believe the current date calculation uses the 7.4% expected rate of return that will be used until the start of next fiscal year, while the July 1, 2017 calculation uses the 10-year Treasury Bill rate plus 2% that will be the permanent variable rate starting next fiscal year.  It would have been preferable if members could enter in their own proposed service purchase date.  This is especially true for anyone who is very close to 20 years of service.  This is because the actuarial formula "punishes" you for buying time that push you up to a milestone like 20 or 25 years of service.  This is why most of us already have been told that we should only buy time very early in our careers or after we have hit a milestone like 20 or 25 years of service.  This is pretty verify with the calculator by comparing the cost of buying one year of service at 19 years of service and at 20 years of service.  (Note: I used my date of birth and an average high salary of $75,000 for all the calculations.  Your numbers will vary based on your own data.)  Today at 19 years of service, a single year will cost $51,576 versus only $21,705 at 20 years, or nearly $30,000 more.  Today at 24 years of service, one year would cost $48,835 versus $27,130 at 25 years of service.  However, the strange thing about the 24/25 yer calculation is that you can buy 0.99 of a year and still get the lower calculation ($21,488) even though you are just short of reaching 25 years.  This does not apply to the 19/20 year calculation where buying 0.99 of a year at 19 years of service will still cost you $51,060.

All this is important for those close to 20 years of service because it makes sense to buy as small amount of time that gets you to 20 years of service.  If you have 19 years of service today, you will still want to buy your service time at the higher discount rate (7.4%) before July 1, 2017 because it will make it cheaper.  However, like on the Price is Right, you will want to purchase it as close to July 1, 2017 as possible without going past that date because then you will be pay less under the aforementioned 19/20 years of service issue.  As an example, say today (August 15) is your 19th year of service, which means you have 10 months, 15 days left in the fiscal year, or 0.874 of a year.  If you were to by that time today, it would cost you $45,062 for the 0.874 year which would get you to 20 years of service.  The other 0.126 of a year would then only cost you $2,735 since you were already vested at 20 years of service.  This would cost you a total of $47,797 for the full year you purchased today.

Now what if you wait until June 1, 2017, one month from the deadline.  You would then be buying only 0.83 of a year to get you to 20 years of  service at the cost of $4,290.  The remaining 0.917 of a year would only cost you $19,903 because you would already have 20 years of service.  The total cost of the one year service purchase on June 1, 2017 would now only be $24,193, a savings of $23,604.  This is a particular example for those really close to 20 years of service, but I suspect the problem will likely affect a quite a few PSPRS members.  Another problem I foresee is with those who may be close to 25 years of service.

As we previously discussed, the actuarial calculations are different for those with the 24/25 years of service issue.  A person in this situation will want to make TWO service purchases before the July 1, 2017 deadline.  As another example, say you have 23 years of service as of today, and you have five years of service time you want to purchase.  What should you do?  You should NOT buy the full five years in one chunk, which would cost you $151,930. (This is again using the same high average salary and birthdate as before.)  You would be better off going to your local PSPRS office and buying 23 months and 15 days worth of time (1.958 of a year) for $42,497, waiting 15 days, then buying the other 3.042 years for $82,531.  This would cost you $125,028 and save you almost $26,902 over buying the whole five years at one time.  Unlike the previous 19/20 years of service example, there appears to be no advantage to waiting closer to the deadline in this situation.  I calculate that waiting would actually cost about $3,000 more to do the two purchases nine months later than today.

As we can see it would be very helpful if PSPRS members could enter the future purchase date.  Of course the further away you are from 20 or 25 years of service, the less helpful this would be since the variables in the calculation, such as your high average salary or the 10-year Treasury Bill rate, are less predictable the farther you are from retirement.  PSPRS members with longer retirement horizons may want to check the payroll deduction option, especially if they have only a small amount of time they want to buy.  For example, a member with five years of service and a high salary average of $50,000 could purchase one year today for $10,540.  Over 60 months, he would pay about $191 every two weeks and spend about $905 in total interest.  Buying more time becomes prohibitively expensive, likely eating up the member's full paycheck, and extending payments out longer increases the interest expense.  Also, it would probably be better to wait and see what inflation is like in the future because, if you are like me and believe it will be higher in the future, it may be cheaper to buy service time in the future as the variable discount rate approaches, equals, or even surpasses PSPRS' expected rate of return.

There may be some difficult decisions to be made if you are planning on purchasing service time, whether via payroll deduction or in a lump sum.  I hope that there will be assistance available to members who will be making important decisions that will cost them thousands of dollars.  Also, assurances need to be received that all requests for service purchases made before July 1, 2017 are honored and calculated correctly.  Members also should not be hit with anymore "surprises" about newly-enforced clauses or last-minute legislative changes.  PSPRS members, past, present, and future, have already been serially shafted by various entities, most egregiously by those whose sole purpose was to protect their benefits at the state level.  Why PSPRS members must have their service purchases discounted at a lower rate than PSPRS discounts its own liabilities is a question that the incurious state public safety unions are unwilling to ask of an indifferent PSPRS administration.  Hopefully, the unions locals, who actually care about public safety workers, will at least be provided the necessary tools to take care of their own members.