Gretchen Morgenson is often a target of heated criticism on the blogosphere, which I have argued more than once is overdone. While her articles on executive compensation and securities litigation are consistently well reported, she has an appetite for the wilder side of finance, and often looks a bit out of her depth. Typically, she simply runs afoul of finance pedants, who jump on misapplication of industry jargon or minor errors when those (admittedly disconcerting) errors fail to derail the thrust of the argument.
But I’ve also seen Morgenson be wrong in the premise of an article. Her weakness is falling for prevailing narratives and it appears that skillful operatives can spin her by making their accounts align with prevailing prejudices. For instance, it isn’t hard to get people to believe an account that depicts Goldman as a bad actor. And while I’m plenty critical of the firm, it really is possible, upon occasion, that Goldman wasn’t the bad guy, particularly when the other party (in this case, AIG) is far from a paragon of virtue. Morgenson jumped on the “Goldman engaged in predatory marks of credit default swaps to bring down AIG” account, a story clearly pushed by AIG employees. The problem was that if you looked at the data (as Tom Adams did) and contemporary accounts (as I had during the crisis) overly generous marks of illiquid paper and complex exposures was widespread (NC had repeated stories on how collateralized loan obligations, less opaque and bespoke than CDOs, were being valued at unrealistically high levels because dealers were stuck with inventory and didn’t want to report losses.). So if you had bothered to do your homework, the assumption would be that AIG and other dealers did have unrealistic marks, and Goldman’s marks were proven to be correct. So if there was a story here at all, it would need to have been a far more nuanced and detailed one than the Goldman-slamming account that Morgenson produced.
I am in no position to say definitively, but Morgenson may have been spun in her report yesterday on a Denver public school financing that turned out worse than initially planned due to financial upheaval. Superficially (and the story presents quite a few general comments and analogies that encourage readers to make the association) this story fits into conventional narrative of “municipality meets banker, banker sells derivative, banker wins and municipality loses, each big time.”
But as with the AIG/Goldman case, an interested party is pushing a story, and while that could easily have been inferred in the AIG/Goldman matter, it isn’t obvious to most readers of the Denver story, which puts them at a disadvantage in assessing it. Michael Bennett, then the school board superintendent, now a US senator, is facing a challenge in the Democratic primaries, and former school board members who are aligned with the opposition are the main (and unnamed) sources for the substantive details in this account. What is particularly sus is the New York Times taking up this story now, shortly before the primaries, when this financing was covered in the Denver media over the past few years.
A buddy of mine (former investment banker turned financial writer, no soft touch, and not a Colorado resident) contends the Morgenson account is off base and forwarded a message from another member of the school board from the time of the financing who lists numerous errors in the Morgenson account. I’m in no position to verify them independently, but the list is long and specific enough to be troubling. The most serious charge is that first, Morgenson got the fundamental impact of the transaction wrong, that even after allowing for its tsuris (and the extra fees involved), the deal still saved the school district money, $20 million to date. In addition, Morgenson claims the school board foolishly entered into a variable rate + swap transaction instead of a “plain vanilla bond,” thus exposing itself to more risk and unnecessary fees. The board member contends that no such “plain vanilla bond” existed for their situation, and the available fixed rate option, after allowing for fees, was more costly than the floater swapped into fixed that the board chose.
The text that follows is from Denver public school board member Jill Conrad:
The Truth of the Matter
I’m all for raising important questions (as I and other board members did in 2008) about major financial and other policy decisions. What disappoints me is the ways in which the facts have been intentionally distorted in many of the blogs and articles I have read on the subject in the last several months. Despite the great reputation (and my own respect for the NYT), I am shocked by the lack of attention to detail and contextual information in the Morgenson article. Here is my take…and an account of at least 18 factual errors in the article…
A. DPS achieved three critical goals with this transaction: we funded the pension in order to allow us to merge with PERA, by merging with PERA we achieved portability, and we added back $20M to schools (previous cuts had been over $80M)
B. Last Friday (7/30) PERA released the CAFR (independent audit) of current funding levels and projected funding: DPS is funded at 88% as of 12/31/09 and will achieve full funding in 2031, 22 years. PERA/school division is funded at 70% with full funding at 2040. Clearly this has had the right effect, financially for DPS employees and academically for today’s kids and teachers.
C. The article is full of inaccuracies and misleading uses of data: I count at least 18 factual errors in this poorly researched and highly slanted article:
1. “Since it struck the deal, the school system has paid $115 million in interest and other fees, at least $25 million more than it originally anticipated.” – Actually, the district has realized $20 million in savings to date compared to not having done the transaction.
2. “While it is possible that the annual costs of the Denver deal will come down in the future, they are now roughly in line with what the school system would have paid in a fixed-rate transaction.” Not true. Fixed rate debt cost 7.25%. DPS is now paying 6.1%
3. “In the end, a deal that JPMorgan said would have an interest rate of around 5 percent spiked to 8.59 percent during its first fiscal year, and has since settled down to an average rate of 7.12 percent today.”
4. To the contrary, the all in cost including all fees of the transaction was projected and budgeted at between 5.75% and 6%, not 5%, at the time the deal was approved. DPS is now paying 6.1% inclusive of all fees, not 7.12% . The cost did spike to 8.59%, inclusive of all fees, during the 2008-09 fiscal year.
5. “And for years, the school system had not met its required annual pension payments to ensure a fully funded plan; by 2007, the school system faced a $400 million gap.” False. With very few and not economically material exceptions the district had made all required annual payments for years. The pension was underfunded because of the 2000 – 2002 market crash.
6. “The Denver schools essentially made the same choice some homeowners make: opting for a variable-rate mortgage that offered lower monthly payments, with the risk that they could rise.” To the contrary, the rate was fixed via the interest rate hedge or “swap” transaction the article later criticizes. The transaction structure is the opposite of a variable rate mortgage where the homeowner is subject to rises and falls in interest rates.
7. “Rather than issue a plain-vanilla bond with a fixed interest rate, Denver followed its bankers’ suggestion and issued so-called pension certificates with a derivative attached. Together, $750 million was raised using the riskier pension certificates.” This is inaccurate and misleading. Regardless of whether the pension debt was fixed or variable rate, the debt would still have been in the form of pension certificates of participation. There was no option of a “plain vanilla bond”. In 1997, we issued fixed rate pension certificates of participation. In 2005 and 2008, we issued variable rate pension certificates of participation, in both cases with a very commonly used interest rate hedge, the “derivative” the article refers to.
8. “To avoid mounting expenses, the Denver schools are looking to renegotiate the deal.” False. Expenses are not mounting. Expenses have declined considerably. Our current all-in costs, including interest and fees, is 6.1%.
9. “And had the school district issued fixed-rate debt, it would not have paid Wall Street the cornucopia of fees embedded in the more complex deal.” Actually, the fees up would have been the same whether the debt was fixed or variable. Ongoing fees were budgeted from the beginning and are all included in calculating the $20 million in savings thus far.
10. “Unlike many school district officials, both men were financially sophisticated and had worked together in the private sector.” False. Bennet and Boasberg have never worked together in the private sector.
11. “Like a homeowner, Denver essentially started out with the equivalent of a standard, fixed-rate mortgage that allowed it to refinance if interest rates fell.” False. Denver’s 1997 fixed-rate pcops could not be refinanced. There were no call provisions in the 1997 pcops or ability to refinance if interest rates fell.
12. “Moreover, refinancing was extremely costly, given the hefty termination fees.” False. There are no termination fees per se. There is a make-whole provision that runs both in favor of and against the banks (and DPS). DPS has the sole option to terminate the swap, which is priced based on publicly quoted Bloomberg rates. If interest rates rise, then the banks owe a termination fee to DPS should DPS choose to terminate. if interest rates fall, then DPS owes a termination fee to the banks. The make-whole fees are exactly the same in either direction.
13. “Agreeing to be locked into a 30-year contract, as public entities have done, is especially costly because getting out of it requires paying penalties to the banks for every remaining year of the transaction.” False. There are no penalties for remaining years of the transaction. There is a make-whole provision that runs both in favor of and against the banks as described in the previous para.
14. “Like the punishing prepayment penalties some homeowners have to come up with when paying off a mortgage early, termination fees on deals like Denver’s are essentially charges levied to rewrite the terms of a contract.” False. There are no charges levied to rewrite the terms of the contract, only the make-whole provision described above. If interest rates rise, the banks owe a termination fee to DPS.
15. “The pension turned in a dismal performance in the credit crisis – as was the case with most such funds – losing almost twice the $400 million borrowed by the school district to plug the pension gap. As a result, the school system’s pension shortfall recently stood at around $386 million, only slightly lower than it was two years ago, and even though $400 million had been funneled into it in 2008.” Article does not mention that absent the $400 million contribution to the pension in 2008, the current pension shortfall would be twice as large as it currently is.
16. “While the pension’s merger with the state system allows Denver’s school system to avoid paying interest on shortfalls, that benefit is temporary. If a shortfall still exists in 2015, the merger requires that it be closed.” False. DPS must continue to pay interest on the shortfall, which is known as a UAAL payment. Article fails to note that Cavanaugh and Macdonald, the independent auditor to PERA, the state pension fund, have projected the DPS pension fund will be 140% funded at the end of the current 30-year project period. The auditor also found that the district’s pension fund was significantly better funded than the statewide division and all of the other school districts, and would be fully funded years before the rest of the school division would be.
17. “Boasberg maintains that the deal has allowed Denver to hire teachers while other school districts are cutting back. But Henry Roman, president of the Denver Classroom Teachers Association, said that fewer teachers had been hired this year than in previous years.” This is remarkably misleading. All school districts in Colorado are facing severe budget cuts. While it is true we are hiring fewer teachers than in previous years, the savings from the pension debt transaction has allowed us to actually hire teachers while most of our neighboring districts are laying teachers off and instituting furlough days.
18. And finally, the article quotes Jeannie Kaplan. It does not attribute the fact that she is a leading fundraiser for Andrew Romanoff, Michael Bennet’s opponent in the primary, and falsely claims that she raised issues about the pension debt in advance of the primary race. Article offers no substantiation that Ms. Kaplan raised this issue in advance of the primary race since no such substantiation exists. The article does not mention that a second board member, Andrea Merida, who has questioned the transaction is a paid staffer for Mr. Romanoff’s campaign.
The fact remains that this transaction has positioned DPS, its employees, and most of all its students to be better off in the long run than they would have been if we had not taken action. I was proud to be a part of the decision then, stand by it now, and am grateful for the leadership and vision of Michael Bennet, Tom Boasberg, and my fellow board colleagues who are focused on one thing-the most important thing-turning DPS around so that all resources, time, and attention serves to drive increases in student achievement, graduation rates, and college success.