The Source
Page 21
But matters can become more complicated, as in the case of swaps. At their simplest, swaps are agreements whereby one party pays a fixed interest rate to a second party; in return, the second party pays a variable interest rate back to the first. That is, two investors exchange risks. The first party is taking on a known risk (fixed rate) while the second is guessing that over time, the variable interest will fluctuate in such a way as to be more profitable than the fixed rate. Because fixed interest rates are known, they are often more expensive—that is, have higher interest rates. For homeowners, getting a mortgage with a fixed interest rate means agreeing to a slightly higher interest rate than that offered by a mortgage with a variable interest rate; the borrower is gambling that interest rates will rise over time and so will accept a slightly higher rate at the beginning to avoid potentially much higher rates over the duration of the mortgage.
This is where swaps come in. If interest rates are particularly high when a borrower needs money, or if the borrower is sitting on a loan that already has a high interest rate, the borrower may want to do something to reduce that rate. Borrowers with a high fixed rate would be interested in a swap that exchanged their high fixed rate for a lower but variable rate. This inherently increases risk. If a city has twenty years of bonds to pay, and each year requires an interest payment of 5 percent (i.e., fixed rate), that city may be interested in swapping with a bank that would pay a variable interest rate that was starting at some lower rate (say, 4 percent) but could potentially rise above the fixed rate over time.
Just as Reagan was transforming the national economy, interest rate swaps were transforming the global economy: private sector investments using swaps rose from $3 billion in 1982 to over $6 trillion in 1993. During the rise of this increasingly creative financing, state governments did not allow local governments to use swaps; all work had to be financed using plain vanilla bonds. But in 1987, both California and Florida passed laws allowing governmental units to use some of these emerging financial tools to maximize their use of existing public funds. That is, to leverage via swaps.
Great savings could be had in using these finance mechanisms: in 1994 the Eastern Municipal Water District in Riverside, California, refinanced its bond debt with an interest swap—and in so doing saved the water district almost $2 million. By 2010, forty states had passed similar laws authorizing municipal authorities to engage in such exchanges, and swaps became an integral part of municipal debt markets; these transactions accounted for an estimated $250–$500 billion. These types of transactions often had their desired effect for the municipalities: they decreased the interest rates municipalities had to pay. Yet there was always a hidden risk. Fixed-rate bonds were secure, and in a sense they separated the finances of municipalities and local governments from the vagaries of the global economy. Interest rate swaps more tightly coupled the sewers with the complexities of Wall Street, and it was unclear whether many local politicians actually understood their implications.23
All of these dynamics were coming together simultaneously at the Cahaba River. In 1986, Jefferson County had the highest rating available for its finances: straight A’s. The county’s first bond issuances in 1997 and 1998, in response to the Clean Water Act requirements for sewage overflows, were for over $600 million, using fixed-rate bonds. As costs for the project began to rise, additional debt was rapidly issued for $950 million and then for another $275 million. This debt was all in forty-year, fixed-rate bonds with interest rates around 5 percent. Long-term, stable, boring, vanilla, municipal finance. Then, in 2002, things changed. The county issued its first variable-rate bonds. They did it again in 2003, for a whopping $2.24 billion. Some of this new finance was used to pay off the previous debt; that is, the fixed-rate bonds issued previously were paid for by a larger amount of variable-rate bonds.
Unfortunately for Jefferson County, the swap it had made and the bonds it used to make the swap were riskier than they realized. In the heady days of the roaring 2000s, the county had assumed it would always be profitable to be connected to global market fluctuations and decided to use an auction process—an auction rate security—to sell its bonds. Typical bond issuance is a singular event; bonds are issued, and various buyers purchase the bonds. Some buy bonds that will mature—be paid back—in five years; others buy bonds that will mature in forty years. In these vanilla bond offerings, the seller knows the interest rates for the entire debt package at the time of the selling, so the city or county knows exactly how much to budget each year for the next forty years. In the approach Jefferson County took, bonds are auctioned at regular intervals—anywhere from every week to every month—to buyers from all over the world. At a car auction, if there are many buyers, the high demand increases the price of the car. But in bond auctions, high demand means that buyers are willing to buy the debt at lower interest rates. If many buyers are interested in Jefferson County’s sewer bonds, the auction may start selling the bonds at 5 percent; but as bidding goes on and there is competition for the bonds, the price goes down, perhaps closing at 3 percent. The more buyers, the lower the interest rate Jefferson County would have to pay. In the early and mid-2000s, the market for bonds was flush, so the interest rate that municipalities could get using an auction was always less than they could get using fixed-rate vanilla bonds. Jefferson County used this auction approach to convert its existing long-term debt with a high interest rate to short-term debt with a lower, variable rate of interest.
Using the auction, however, meant that Jefferson County now had significant debt in the form of variable interest rates, which made the county nervous. So, it went to its bank, J.P. Morgan, and swapped interest rates: The county paid $25 million up front to J.P. Morgan for the ability to pay the bank a fixed interest rate of 3.678 percent. In exchange, J.P. Morgan agreed to pay Jefferson County a variable interest rate pegged to a global bond index. Interest rates had been inordinately low for years, and the county naturally assumed, per J.P. Morgan’s counsel, that interest rates would inevitably rise. If they did rise, J.P. Morgan’s payment to the county would exceed the amount the county had to pay the bank and thus generate revenue for the county. All of this made perfect sense for the county as long as two things happened: investors kept buying bonds, and global interest rates started to rise.
What happened if there were fewer and fewer buyers? In that then unheard of worst-case scenario, the lack of sale would trigger the bonds to be sold at some very high fixed rate—perhaps as high as 15 percent. As a safeguard to prevent the auction from entirely failing or the rates from going too high, Jefferson County relied on its underwriting bank, J.P. Morgan Chase, to step in and buy the bonds at an agreed-upon maximum rate if there were no other buyers when the auction was set to close. Of course, that agreement assumed the bank had money.
By 2005, more than 25 percent of outstanding municipal debt in the United States was locked into auction rate securities and interest rate swaps. Local governments, like Jefferson County, were leveraging their infrastructure systems as capital to generate “synthetic revenue” through increasingly complex financial markets. All of this debt, however, tied local revenue and spending to global credit markets, in the latest shift in U.S. government finance. Water infrastructure finance had gone from state and local dominance in the nineteenth century to federal dominance in the twentieth and back to local just before the turn of the twenty-first century.
At the beginning of the twenty-first century, global finance played the central role. The use of credit swaps to generate synthetic revenue also signaled a significant philosophical transformation: no longer was wastewater infrastructure a public service, to be funded solely through tolls or property taxes in the region associated with that service. Now credit swaps were generating revenue, and entrepreneurial cities were using infrastructure as their collateral for entering the global finance market. During the first five years of the twenty-first century, Jefferson County had grown disturbingly entrepreneurial. The county’s sewer system debt had
grown to over $3 billion, and the county began using swaps even more aggressively; between 2002 and 2004 alone, the county entered seventeen swap agreements with a total value exceeding $5.8 billion. That is, the swaps were more valuable than the bonds they were being used to hedge. With almost $6 billion in swaps, Jefferson County (population 660,000) had only $1 billion less in swaps than the entire state of New York (population 19 million). Far from using swaps to hedge fluctuations in interest rate, the county was using them to speculate on those fluctuations. Its chips were all in. The county was completely dependent on global interest rates rising and auction prices falling.24
In January 2008, both Standard & Poor’s and Moody’s gave the county their highest credit rating. Only a few days later, on Valentine’s Day 2008, the roof caved in. The global market froze as the housing market collapsed at the opening of the Great Recession. The debt auction went silent.
With no buyers in the auction at the expected rates of around 3–4 percent, the default base interest offering jumped to 10 percent. The new rate added $700,000 per week in interest payments alone to Jefferson County’s expenses, which it was nowhere close to being able to pay. At the same time, alarmed by the global market freeze, the federal government infused enormous amounts of money into the financial system, causing the global bond index, ironically, to fall. As the index on interest rates dropped, the interest rate swap with J.P. Morgan backfired: Jefferson County had to continue paying the same fixed amount to the bank each month, but the amount that the bank had to pay the county plummeted. The last line of defense for Jefferson County was that their bonds were insured. But their insurers were also in the business of insuring home mortgages. As the subprime mortgage crisis hit, the bond insurance companies disappeared—along with any new bond buyers for the auctions. Every assumption Jefferson County had made about the financial system had been fundamentally wrong. Only eight days after the rating agencies had given Jefferson County sewer bonds straight A’s, both S&P and Moody’s gave the bonds junk status.
Jefferson County stumbled along for a few years before finally filing for bankruptcy in November 2011. The Securities and Exchange Commission (SEC) sued J.P. Morgan, which paid a penalty of $25 million to the SEC and $50 million back to Jefferson County, and then walked away from nearly $650 million in swap cancellation fees that the county was supposed to pay the bank. By 2011, several county employees and former contractors had been indicted by federal prosecutors in connection with the sewer project’s finances for corruption and bribery. Despite all of this shady dealing between the county and Wall Street, the county had to provide the still-necessary funding for wastewater infrastructure as well as its enormous financial miscalculations; residents of Jefferson County saw their sewer rates triple, taking the average sewer bill from about $60 per month to almost $200.25
Water pollution is expensive. So much money is needed to deal effectively with sewage that financing the necessary infrastructure is inseparable from larger questions of how the entire government generates and spends revenue. This reality of sewers and polluted rivers shows that ideological, political battles have fiscal realities ranging from Chicago innovating the “special economic district” to FDR unleashing unheard-of sums of federal funds for treatment plants to Jefferson County using its sewer system to finance absurdly complex financial instruments. Throughout American history, paying for sewers has resulted in tectonic shifts in the political and financial structures.
Why is it that sewers are often at the cutting edge in finance? Because having clean water in a developed country is a nonnegotiable expectation. A government is not considered competent if its citizens cannot drink the publicly provided water without dying. And since the mid-1900s, a government is not considered competent if the rivers and streams under that government are little more than flowing cesspools. These expectations create a nationwide, perpetual demand for water and sewer projects, and the projects are inordinately expensive. As a nation we have not just developed new ways to keep our water clean but also, by necessity, developed new ways to finance those efforts. Thus innovation in governing rivers goes hand in hand with innovation in how we finance public investment.
PART FOUR
REGULATION
CHAPTER 9
Regulating Power
On January 23, 1979, a group of federal bureaucrats sat around a table to decide whether a species of fish should continue to exist. The meeting started at 9:00 in the morning, and by 9:45 the group had voted unanimously to kill a dam and save the fish.1
This dam-killing committee was formed to help reconcile the conflicted American ideas of what the federal government’s role in managing—regulating—society and the economy should be. For over two centuries the power produced by rivers via dams was the backbone of America’s industries, from mills grinding flour for colonial villages to factories manufacturing weapons during World Wars I and II. The centrality of power to society required America to grapple with the relationship between private enterprise and government: How would America regulate something that society depended on?
Regulation is used by government to restrict individual rights and property for the common welfare. Every government must negotiate how to encourage economic and technological development while also ensuring public safety and a certain amount of stability and competition in the economy. Regulations, and the rights they affect, are therefore not static: as industries, technologies, and society evolve, the range of rights considered sacrosanct and those the government deems necessary to regulate also shifts. In the eighteenth century, government had to regulate the bread bakers and innkeepers; in the late twentieth century, it was telephone and pharmaceutical companies; in the twenty-first century, it is drones and the Internet.2
Among the aspects of the economy and society that the United States has regulated over the years, hydropower has had a disproportionate effect in shaping the government’s ideas about how to approach regulation. Dams have played a central role in many societies, but their impact on American ideas of regulation has been particularly strong; the United States came into existence—and grew in geographic and economic scope—in step with the industrial revolution. Just as America was developing its political ideology, the industrial revolution was changing how power was used—in ways that once were unimaginable. And while England’s industrialization was powered by coal and steam, America’s ran on dams.
To appreciate how essential dams were in the nineteenth century, simply look at the 1840 U.S. Census: It found that almost every river had a dam, and many rivers had dozens. In total, the twenty-six states that made up the United States at the time had around 65,000 dams. With a population of only 17 million at that time, the United States had one dam for every 261 people.
One of the highest densities of dams in the young country was found on New England’s Merrimack River, which culminated in a series of dams in Lowell, Massachusetts. Whereas cities like New Orleans were built by draining land and building on top of whatever dry patches could be scratched out, and cities like Los Angeles or Phoenix had to bring water into a city through great canals and irrigation ditches, cities on the Eastern Seaboard—like Lowell—were built on a series of engineered miniature waterfalls.
Lowell is located at the confluence of the Merrimack and Concord Rivers, just downstream of Pawtucket Falls, where the Merrimack drops through 30 feet of cataracts—the Fall Line. Timber was the predominant commodity on the upper Merrimack River, and loggers moved this commodity in vast rafts down the river. At Pawtucket Falls these rafts had to be halted, broken up and hauled by oxen over land around the falls, and then reassembled into rafts to be floated on downstream to the coast. As at most other significant falls along the rivers of the Eastern Seaboard, before long a small company built a lock and canal to get boats and rafts of timber around the falls more easily.
Even though the Fall Line was a chronic barrier to navigation, it was also a source of power. Both at the Fall Line and in the mountains upstream, rivers
plunged through the sort of rapids and moderate waterfalls that were rare in England but ubiquitous in America. Heavy precipitation along the Eastern Seaboard created strong water flow that, when combined with the steep vertical drop to the coast, made the rivers ideally suited for turning waterwheels, the leading technology of the time. Colonists all along the East Coast initially put waterwheels to work in mills to process timber, which was essential for building settlements and one of the key raw materials that was plentiful in America but in short supply in Europe. By the time the earliest sawmills were built in England in the 1660s, several hundred were already being used in colonial New England. The colonists processed some timber for lumber export to England, but they used much of it for building their own settlements.3
Also ubiquitous around the Fall Line were gristmills, indispensable to colonists for grinding wheat to flour or corn to meal. Without a nearby mill, settlers had limited options: they could either transport their grain overland to the closest settlement with a mill for grinding, or they could grind it themselves by hand or with their limited livestock. Either option was time and labor consumptive. Grinding a single bushel of wheat into flour required about two full days of human labor, or a few hours for a horse-drawn mill. Contrast that with the typical eighteenth-century water-powered gristmill, which could grind dozens of bushels of flour or cornmeal per day. Being located near a mill offered settlers relief from some of their most demanding tasks.4