Comptroller's Pension Report 2000-2010

On April 6, 2011, Comptroller Liu released a report analyzing NYC pension cost over the last decade. This is generally a good report. It is a more complete look at the NYC pension problem than what most commentators produce. Liu has his agenda like everyone else but he is way out in front on content and balance.

The starting year used in the report is 1986. That was the year Harrison Goldin started submitting city financial reports to the GFOA. At the time, I thought it was a PR exercise but it was a major step towards transparency for city and pension financial records.

The heart of the report, on page 2, focuses on what Liu considers the most relevant causes of the NYC pension underfunding problem. I have listed them below along their FY-2010 increased cost:

1. lower investment returns - $3.1B
2. benefit enhancements put in place in 2000 - $2.1B
3. actuarial losses and revisions - $790M
4. benefit enhancements put in place after 2000 - $264M
5. higher than expected investment and administrative fees - $313M.

Lower Investment Returns

The report correctly identifies the prime cause of underfunding as the market collapse since March, 2000. But on this key point, the report fails to add the fact that the pension fund trustees exacerbated the market collapse by poor investment decisions.

The actual closing balance as June 30, 2010 of the five city pension funds was $89.9B. In comparison, the closing balance as of June 30, 2000 was $105.6B, a decrease of $15.7B over ten years. It is also distrubing that approximately 10% of the 89.9B is in illiquid limited partnerships whose values are only an estimates.

The employer pension contributions for the 11 years from 2000 to 2010 were $42.9B. The employee contributions for these same 11 years were $7.4B.

The pension contributions (city only) for FY-2010 were 11.2% of the total city budget ($6.651B vs. $59.479B). The total for all employers for all five pension funds was $7.765B for FY-2010.

With the $42.9B contributions, if the funds had earned 8% each year during this period, the closing balance would be $168.8B. No one would be discussing any pension crisis.

In his simulation for the five pension funds, the Comptroller estimated the June 30, 2010 closing balance at $139.2B. Rather than the actual $42.9B, Liu used a much lower level of employer contributions, $11.8B, for the 11 year period from 2000 to 2010.

As of June 30, 2010, the NYCERS actuary, Bob North, estimated (using the EAN method) the pension liabilities for the five pension funds at $145.8B. This generates a short fall of $56B ($145.8B - $89.9B).

Using a more traditional 4.5% of the total city budget for employer contributions ($28.1B), I estimated that the 8% June 30, 2010 closing balance for the five pension funds would have been $151.6B.

The $151.6B figure would easily cover North’s estimate of pension liabilities even with the increased benefit structures outlined in the report. You, therefore, can make a plausible argument that the investment collapse caused the entire problem.

Bad Investment Decisions

But the killer point is that as of the summer of 2002 all parties knew that 8% was an irrational investment target.

But because of the trustees’ adherence to the 8% target, (see my March 15 posting), the pension funds continued to follow their 70%/30% investment strategy after the 2000-2003 market collapse and even increased the risk level with an new 11% commitment to private equity and real estate. This was motivated by the desire to avoid paying the higher employer contributions that would have been required with a lower more realistic target rate. In the end, they wound up paying more and getting less. They lost on both ends.

The impact of the collapsing markets was magnified by the 8% decision. In my March 15 posting, I stated that if NYCERS had adopted a conservative strategy from 2005 to 2010, it would have increased the its closing balance for 2010 from $35.4B to $42.8B. Projecting this number to all 5 systems, the closing balance could have been $108.7B instead of the $89.9B, an $18.8B increase.

It is unrealistic for us to expect the trustees to admit such a huge mistake. So they blame all the loss on the market collapse and accept no responsibility. They do, however, regularly claim credit for investment increases.

The actual flat rate of return for the 5 city pension funds for 2000-2010 was 3.85%. You can see the specific rates of returns for 2001 to 2010 on page 6. In contrast, NYCERS 10 year rate of return on its government bonds was 7.72% as of June 30, 2010. It is easy to see the impact of aggressive investment decisions. There are bad decisions in the investment world.

Note: The 2000 market value reset issue, see page 4, is important but rarely mentioned. In a classic case of bad timing, March, 2000 was the start of the dot.com market collapse. On page 5 you can see how the level of employer contributions (1983 to 1999) benefited from the long bull market and the pension benefit reforms under Tier 3 & 4. The sharp drop, however, in the 2000 employer contributions was due to the market restart.

Benefit Enhancements Put in Place in 2000

This is a very comprehensive breakdown of the pension benefit improvements from 2000 to 2008. See pages 18 & 19. This is the most intriguing part of the report. It attempts to quantify the effects of the benefit enhancements. It catalogues and presents purported annual costs for individual benefit enhancements. It puts the benefits in a cost framework. I am, however, skeptical of the accuracy of the cost figures. Assuming that they are accurate, it provides a guideline for corrective action with respect to the benefit structure.

I do, however, particularly question the estimated cost for the city of the state wide COLA benefit enacted in 2000. Liu claims that this benefit costs the city $1.373B in FY-2010. I doubt that the cost is that high.

For this benefit, the city has an exposure to about 180,000 pensioners.

1. NYCERS: 72,000
2. TRS: 79,000
3. BERS: 10,000
4. Police: 8,000
5. Fire: 11,000

Generally at age 62, the annual COLA on average starts at $300 and grows by $300 per year per pensioner for his/her remaining lifetime. This benefit conservatively has a life cycle on average of 18 years.

This translates into an average annual cost of a $2,250 per pensioner or a $486M ($300 * 18 / 2 * 180,000) total annual cost on a pay as you go basis. This benefit does not grow after it reaches a steady state.

This means that the $1.373B cost for FY-2010 seems to be way off track, even considering an effort for future funding.

This type of discrepency means that the benefit costs have to be more carefully researched with extensive documentation supporting the estimated costs. .

Actuarial Losses and Revisions

Actuarial profits and losses occur when the actuary makes mistakes in his assumptions. If he/she is too conservative, then you have a profit. If he/she is too aggressive, then you have a loss. I don't think this is really a cause of increased pension costs but reflects a bias on the actuary's part to reduce pension costs.

Benefit Enhancements Put in Place After 2000

What is interesting about this point is why has the current mayor agreed to give any benefit improvements during this period. He has also given significant salary increases during his nine years in office in spite of the huge pension overhang that has existed since 2002. In particular, the two 4% increases given to DC-37 in the fall of 2008, concurrent with the fall of Lehman, are disturbing at best.

Higher Than Expected Investment and Administrative Fees

This is a valid area of concern which Liu is presenting using the two year lag.

On an accrual basis, the five systems incurred investment costs of $101.9M for FY-2002. The FY-2010 cost was $462.8M, a 454% increase. I did not start with FY-2000 because prior year’s investment costs were bundled along with securities lending costs and were not broken out in the CAFR reports.I have previously commented on the insanity gripping the trustees concerning investment fees.

Two of the systems, NYCERS and TRS, incurred administrative costs of $52.9M for FY-2000. The FY-2010 figure is $115.7M. NYPPF started incurring costs in FY-2002($8M) and BERS($4M) in FY-2003. Liu, quite correctly, allows for 3% per annum increase which would have produced a FY-2010 cost of $86M. The actual cost of $115.7M reflects a 6.3% average annual increase for the 10 years, way above the rate of inflation.

Both of these costs must repaid by the employers two years later with a 8% annual interest charge. For example the $115.7M admin charge for FY-2010 must be paid in FY-2012 with interest totaling $134.9M. The $462.8M charge must paid back in FY-2012 at $539.8M. These costs are directly under the control of the trustees. Why are trustees letting them costs run wild?

Liu attempts to compare these cost to other public pension funds. That is a waste of time. If everyone is jumping of the cliff, are you going to jump off the cliff too?

Closing

There is an excellent closing to the report on page 12. It calls for increasing investment income while reducing volatility. It, however, fails to own up to the failings of the trustees with respect to past investment decisions and to the exploding investment and administrative costs that the trustees have immediate control over.

There is a grudging consensus that the pension benefit structure needs to be cut back at the city, at least to the Tier 5 level at the state. Most of the benefit reform is already in place. Newly hired city police officers and firefighters are under Tier3 and newly hired teachers are under Tier 5. Newly hired general city workers will have to moved to Tier 5 in the same manner as general state workers.

But this benefit reform must be coupled with a strategey that adopts a more prudent, conservative and effective investment plan and outlaws all campaign contributions to any pension fund trustee from any contractor who is work for funds.