The Philadelphia school district is the eighth-largest in the country, with a $2.3 billion operating budget for 242 schools serving 150,000 children, over 80% of whom are poor. The finances of the district, along with the rest of the city, are under a lot of pressure these days. That, of course, makes them no different from a lot of urban school districts. With state aid unable (or unwilling) to keep up with their expenses, and local property taxes pushed up as high as politically possible, the department has been in dire straits for some years.
From 2002 until 2007, the city executed a series of “interest rate swap” agreements with Wall Street banks, including Wells Fargo, Morgan Stanley, Citigroup and Goldman Sachs, to transform their floating-rate debt into fixed-rate debt. For a fee, the Philadelphia school district agreed to pay the dollar value of the fixed-rate obligations of the banks, and in exchange those banks agreed to pay the floating-rate obligations of the school district. The intent was to make budgeting more predictable, and possibly to save some money. Accounts differ at this point, since things did not turn out well. Unfortunately, after the agreements were in place, interest rates plunged in the aftermath of the 2008 financial crisis, and they remain at historic low levels today, five years later. Philadelphia’s payments to the banks at the fixed rates didn’t plunge, but the payments the banks made to Philadelphia went down to near zero. Suddenly, what seemed like a bright idea at the time had become a disaster, and as of 2013, the school district and the city have lost $331 million in these deals, including interest rate payments and more than $110 million in cancellation fees. They remain on the hook for hundreds of millions more, and the banks have been utterly unwilling to forgive or renegotiate these deals. As of July 2013, the district has plans to close dozens of schools and lay off thousands of employees to deal with their ongoing fiscal crisis.
It’s not an exaggeration to say that the landscape of the financial markets these days is a veritable minefield for taxpayers. Governments need banks, and bankers know it—and frequently use the fact to their advantage. This, of course, is no different than any other supplier of services to a government—school districts need teachers, and teachers know it, too—until it goes over the line, past what used to be the bounds of propriety. In the case of Philadelphia’s interest rate swaps, the abrupt decline in interest rates made what had seemed a prudent hedge, beneficial to both parties, into a phenomenally profitable deal for just one. Imagine a wealthy banker offering his neighbor some money for his cabin on a cruise because it’s a little closer to the lifeboats. That’s paying to hedge some risk. He’s happy; the neighbor’s happy. There’s a mutual benefit: the banker has reduced his risk, the neighbor has a little extra money. Now imagine the ship has sunk, the neighbor is in the water, swimming towards the lifeboat. The banker in the boat extends an oar to the swimmer. . . and clubs him on the head with it. “He would have sunk our lifeboat,” the banker says, as the neighbor drifts away. That’s not hedging, it’s horrifying. The fact that the banks involved are determined to keep all the profit, at the expense of Philadelphia school children, is an appalling position to take, comparable to the banker already in the lifeboat claiming it would have been too risky to give up his life vest to save the neighbor. “A deal’s a deal,” they’ll say. It is merely an assumption, but it seems a safe one, that none of the bankers involved have children in the Philadelphia schools.
Actions like this should cause bankers who made the Philadelphia deals to be shunned in polite society. But it won’t: welfare fraud is frequently punished by jail, but bankers taking advantage of governments are punished with big bonuses.
This is barely the beginning. Financial transactions involving the government are often large and complex. A bond sale can involve a small legion of bankers and lawyers, and take several steps to be executed. This provides a multitude of nooks and crannies into which fees and profit can be tucked, especially where there is no tradition of fiduciary responsibility of the banker to his or her customer. There are also many transactions whose purpose involves the transfer of risk from a bank to a customer. Usually for a fee, of course, but risk is hard to quantify and hard to communicate effectively, even when one wants to. Far too often, our governments have been the victims of these transactions, not the beneficiaries.
Here, then, is a short and incomplete list of the ways in which banks profit in unseemly fashion from their government customers. There are undoubtedly many more, yet to be uncovered.
- Swaption Philadelphia’s interest-rate swap was bad enough, but in 2004, Northampton County, a rural county in eastern Pennsylvania’s Lehigh Valley, sold an interest rate “swaption” to Merrill Lynch (now Bank of America). A swaption is a promise to enter into an interest rate swap, and in this case, Merrill Lynch paid the county about $1.9 million up front (less $300,000 in fees) for the right to swap fixed for floating interest rates in 2012. By 2012, it was clear this was a really bad idea, but terminating the agreement on the eve of the swap invoked penalty clauses that cost the county $27 million.
- Bond churning In 2005, the state of Louisiana sold $650 million of bonds to refinance existing debt at a lower interest rate. Of course, $45 million of that debt had only been issued three months earlier, in October, at more or less the same interest rate. Citigroup, the underwriter of the earlier bond, and one of the underwriters of the bigger sale, essentially sold the same bond twice, earning $665,000 in fees for the first sale, and $104,000 in fees for refinancing just those October bonds.
- Letters of credit Municipal bond borrowers frequently use a bank’s letter of credit to prop up their own bond rating and lower the interest to be paid on a bond. Essentially the bank is offering a guarantee for the bond, and so its bond rating becomes more material than the municipality or agency issuing the bond. The problem is that letters of credit expire, typically in three or four years, while a bond might not be paid off for 30 years. When the letter of credit expires, the municipality must get another, or be forced to pay the bond back immediately. Unfortunately, with the expiration of several large banks in the 2008 crisis, letters of credit are harder to come by now, and the fees banks charge for them have risen dramatically. In 2010, the Port of Oakland saw a $2 million increase in their LOC fees to back up a $200 million line of credit using the commercial paper market. (These numbers are from the Port of Oakland’s annual reports and agendas. On the brighter side, in 2012, with competition in the LOC market somewhat restored, Oakland’s LOC fees have come down somewhat.)
- Pension obligation bonds The idea with a pension obligation bond is to borrow at a low rate, put the resulting money into the pension funds, and convert a 7.5–8.5% debt on the pension unfunded liability into a 5–6% debt, more what a municipality might expect to pay on a taxable bond. This might sound like a great idea, but in the end it’s just gambling, since the 7.5–8.5% target on pension investments is the discount rate for the obligations, which is ever so slightly different from the expected rate of return. That is, everyone fondly hopes the fund managers will meet that target, and the evidence is that they do a decent job of it on average, but there is no guarantee. In 2008, Connecticut borrowed $2.28 billion to top up its teachers retirement fund. That would be roughly when the equity market, in which most of those funds were invested, tanked. When the bond was issued in April, the Dow Jones average stood at 13,000, and by the following November, it was just over 6,600. At this point, in 2013, the investments may have returned to near their original value, but Connecticut has been paying 5.88% interest in the meantime, along with ample fees to the banks that underwrote the bonds in the first place. There are years left on the terms to those bonds, so there is time to catch up, but the risks are immense. (Connecticut is holding the course.)
- Capital appreciation bonds A capital appreciation bond is a bond that only makes one payment. It’s a new thing, related to an older version called a “zero-coupon” bond. A zero-coupon bond is sold at a discount from the face value, while a CAB is sold as returning a fixed rate of interest. In both of them, the value of the bond plus its interest is all paid at the end. The important difference from a government’s point of view is that the face value of the capital appreciation bond is the discounted number, so if you’re worried about the total debt of your town, this can make it look smaller. For example, you might sell a $1,000 zero-coupon bond with a ten-year term for $558 to someone who bids a 6% interest rate. The bond will count as a $1,000 debt in the municipality’s accounting. A capital appreciation bond of the same amount would count as a $558 debt in the accounting. Even though it’s a great way for a mayor to hide how much debt had been taken on, this doesn’t sound terrible. Unfortunately, the unpaid interest compounds if you don’t make interest payments, so the costs add up quickly. You’re paying interest on the interest. Starting at $558, a ten year run of semi-annual payments at 6% simple interest would wind up costing $750 total, or about $192 in interest, less than half the $442 interest cost of the capital appreciation bond.For a longer term, it’s pretty easy for the interest to exceed the principal by a lot.In 2011, the Poway Unified School District which serves the city of Poway, California, and part of San Diego, borrowed $105 million on a 40-year capital appreciation bond. They will make some interest payments starting around the year 2031, but otherwise no payments are due until maturity, in 2051. When the bond is done, they’ll have repaid the principal and $877 million in interest, eight times the principal.
- Bond flipping There is a conflict in incentives in the bond underwriting business. The issuer of the bond wants as low an interest rate as possible (as high a price as possible at the initial bond sale), but the underwriter wants to stay friends with its big institutional clients, who want to purchase the bonds as cheaply as possible. And a big part of the reason they want them cheap is to resell them. If a bond dealer or big institutional investor can buy a bunch of newly-issued bonds from some city at 5% and then turn around and sell them at 4.5%, he has made a tidy profit, but he’s also proven that there are people who would have bought them from the city at 4.5%. The higher interest rate is money paid by taxpayers, not going to a public purpose, but only to make a profit for the underwriter’s friends. Banks have been accused of giving preference to certain dealers during bond sales, filling their orders at low prices while there are still other orders unfilled. The SEC and FINRA (a financial industry trade group that moonlights as an industry regulator) have discussed new rules to prevent this in the past couple of years, but the obscurity of the bond sale process has prevented any real oversight.
- Padding bond sales It is very difficult to get more than a vague picture of the underwriting profit earned from an issuer for any specific bond transaction. The bond prospectus is supposed to answer such questions, and some of the fees may be outlined, but to understand the amount earned usually requires understanding the whole transaction. This makes it relatively easy for banks to include costs that have nothing to do with the bond. In December 2012, FINRA fined five banks $3.3 million for padding the fees they collected from bond sales with dues they paid to Cal PSA, a California lobbying association. The banks, including Citigroup, Goldman Sachs, JP Morgan, Merrill Lynch (BankAmerica), and Morgan Stanley, were also forced to pay $1.13 million in restitution to a number of municipal and state issuers in California. Cal PSA was billing its members according to the volume of bonds they underwrote, so apparently members found it easy just to tack the bills onto their underwriting fees.
- Tax Anticipation Notes Tax revenues have an ebb and a flow each year. Income taxes tend to bump in April, despite the withholding rules, and sales tax collections bump in December. A government that depends on income taxes might find itself short in March, and can sell a “Tax Anticipation Note” (TAN) to get through the month until the April tax returns start rolling in. (You’ll also see them as “Revenue Anticipation Note.”) In an earlier, more innocent time, the chance of a shortfall would be the reason a government would maintain a cash reserve. This reserve would be idle most of the year, when it could be invested in T-bills, and only used for a short time. A government that relies on TANs not only gives up that investment income, but also pays all the bond issuance costs each year. San Diego pays about $200,000 each year on the TAN issuance fees alone. This year, they expect to pay $300,000 more in interest, down from $5 million in 2008.
- LIBOR rigging The London interbank interest rate (LIBOR) is a benchmark interest rate, to which millions of other interest rates in hundreds of trillions in investments refer. When a group of banks admitted to rigging LIBOR in 2012, they were admitting to having stolen money from the parties involved in these agreements. Some $200 billion of municipal interest rate swaps sold before the 2008 crisis were tied to LIBOR, and rigging the rate made the swaps themselves more expensive and getting out of them more expensive, too. Bloomberg news estimates that the additional cost to municipalities exceeds $6 billion.
In addition to all these, there are the more pedestrian sorts of fees, the sort that plague any modern bank customer. Because many government bank accounts have very high traffic, the fee income can be quite substantial.
These incidents are nothing more than symptoms of a larger situation. School districts, cities, and counties across the country have been squeezed, and squeezed hard, by the anti-tax movement of the past 35 years. Philadelphia’s schools were looking for security with the interest rate swaps, but they were also eager for the fees the banks paid to engage in them. Without adequate support from the state, school district officials were forced to “think outside of the box” and this is what they came up with.
In the same way, the reason governments don’t keep a cash reserve is that their budgets have been depleted by years of squeezing. In a climate of interest rates as low as they are today, covering a cash shortfall through TANs isn’t that big a deal, but interest rates aren’t going to stay this low forever, and rebuilding those reserves when interest rates rise will be quite painful, if it’s possible at all. In other words, these governments are victims of predatory public finance, but the reason they became prey is because of the squeezing.
What’s especially remarkable about the conduct on display here is not simply the lack of consideration for schoolchildren, taxpayers, and the rest of the 99%, but that despite all the squeezing, many governments, if not most, have the assets they need to conduct their own financial business. In 2012, San Diego issued $101 million in TANs, but had over $2 billion in cash and investments on hand, plus another $7 billion in investments they are managing for others, including their pension system. Approximately $1.2 billion of it is invested in short-term loans to corporations and other governments. San Diego’s ends could be handled with their own means.
Unfortunately, our governments’ wealth is seldom organized in such a way that they can do it easily, and there exists a legion of financiers to say they can’t. The work before all of us now is to consider ways to create a better financial system, one that works for all of us, not just the one percent.
Some of this work is already underway. For example, in 2012, the Build America Mutual bond insurance company was formed to insure municipal bonds. It is a mutual insurance company, so is owned by the municipalities whose bonds it insures. It is also brand-new, so whether it can maintain a level of service and accountability better than its for-profit competitors is an open question, presumably to be answered over the next few years. Over the past couple of years, cities and counties across the country have been exploring the possibilities of establishing financial institutions and banks to serve their own financial needs, to focus lending within their borders, or just to move their money into banks that care about them. The state of North Dakota does this, why not the city of San Diego?
Other small reforms are feasible. Simply creating a legal requirement that banks act in a fiduciary capacity for its customers would go a long way. For example, in Vermont, mortgage brokers can be legally liable if a customer’s loan goes bad due to fraud or inadequate disclosure. Consequently, Vermont has seen virtually no increase in foreclosures since the 2008 crisis, while the rest of the country has been inundated. Sometimes simply being clear about exactly who is on the hook when things go bad is reform enough.
It is crystal clear that in many ways, banks regard our state and local governments as marks to whom they can sell the newest, shiniest, financial gimmick, no matter how risky or rigged. The good news about this is that state and local governments can be fairly accessible, and even occasionally responsive to public pressure. Our cities, towns, counties, and states have the financial wherewithal to avoid being preyed upon, but it seems we need to force them to act that way.
[This is adapted from the forthcoming book, "Checking the Banks", by Tom Sgouros, a primer about banking mechanics, possible reforms, and government finance, about which you can read more at lightpublications.com/checking]