A decade ago, as the financial crisis raged, America’s banks were in ruins. Lehman Brothers, the storied 158-year-old investment house, collapsed into bankruptcy in mid-September 2008. Six months earlier, Bear Stearns, its competitor, had required a government-engineered rescue to avert the same outcome. By October, two of the nation’s largest commercial banks, Citigroup and Bank of America, needed their own government-tailored bailouts to escape failure. Smaller but still-sizable banks, such as Washington Mutual and IndyMac, died.
After the crisis, the goal was to make banks safer. The 2010 Dodd-Frank law, coupled with independent regulatory initiatives led by the Federal Reserve and other bank overseers, severely tightened banks’ ability to engage in speculative ventures, such as investing directly in hedge funds or buying and selling securities for short-term gain. The new regime made them hold more reserves, too, to backstop lending.
Yet the financial system isn’t just banks. Over the last ten years, a plethora of “nonbank” lenders, or “shadow banks”—ranging from publicly traded investment funds that purchase debt to private-equity firms loaning to companies for mergers or expansions—have expanded their presence in the financial system, and thus in the U.S. and global economies. Banks may have tighter lending standards today, but many of these other entities loosened them up. One consequence: despite a supposed crackdown on risky finance, American and global debt has climbed to an all-time high.
Banks remain hugely important, of course, but the potential for a sudden, 2008-like seizure in global credit markets increasingly lies beyond traditional banking. In 2008, government officials at least knew which institutions to rescue to avoid global economic paralysis. Next time, they may be chasing shadows.
The 2008 financial crisis vaporized 8.8 million American jobs, triggered 8 million house foreclosures, and still roils global politics. Many commentators blamed a proliferation of complex financial instruments as the primary reason for the meltdown. Notoriously, financiers had taken subprime “teaser”-rate mortgages and other low-quality loans and bundled them into opaque financial securities, such as “collateralized debt obligations,” which proved exceedingly hard for even sophisticated investors, such as the overseas banks that purchased many of them, to understand. When it turned out that some of the securities contained lots of defaulting loans—as Americans who never were financially secure enough to purchase homes struggled to pay housing debt—no one could figure out where, exactly, the bad debt was buried (many places, it turned out). Global panic ensued.
The “shadow-financing” industry played a role in the crisis, too. Many nonbank mortgage lenders had sold these bundled loans to banks, so as to make yet more bundled loans. But the locus of the 2008 crisis was traditional banks. Firms such as Citibank and Lehman had kept tens of billions of dollars of such debt and related derivative instruments on their books, and investors feared (correctly, in Lehman’s case) that future losses from these soured loans would force the institutions themselves into default, wiping out shareholders and costing bondholders money.
The ultimate cause of the crisis, however, wasn’t complex at all: a massive increase in debt, with too little capital behind it. Recall how a bank works. Like people, banks have assets and liabilities. For a person, a house or retirement account is an asset and the money he owes is a liability. A bank’s assets include the loans that it has made to customers—whether directly, in a mortgage, or indirectly, in purchasing a mortgage-backed bond. Loans and bonds are bank assets because, when all goes well, the bank collects money from them: the interest and principal that borrowers pay monthly on their mortgage, for example. A bank’s liabilities, by contrast, include the money it has borrowed from outside investors and depositors. When a customer keeps his money in the bank for safekeeping, he effectively lends it money; global investors who purchase a bank’s bonds are also lending to it. The goal, for firms as well as people, is for the worth of assets to exceed liabilities. A bank charges higher interest rates on the loans that it makes than the rates it pays to depositors and investors, so that it can turn a profit—again, when all goes well.
When the economy tanks, this system runs into two problems. First, a bank’s asset values start to fall as more people find themselves unable to pay off their mortgage or credit-card debt. Yet the bank still must repay its own debt. If the value of a bank’s assets sinks below its liabilities, the bank is effectively insolvent. To lessen this risk, regulators demand that banks hold some money in reserve: capital. Theoretically, a bank with capital equal to 10 percent of its assets could watch those assets decline in value by 10 percent without insolvency looming.
Yet investors would frown on such a thin margin, and that highlights the second problem: illiquidity. A bank might have sufficient capital to cover its losses, but if depositors and other lenders don’t agree, they may rush to take their money out—money that the bank can’t immediately provide because it has locked up the funds in long-term loans, including mortgages. During a liquidity “run,” solvent banks can turn to the Federal Reserve for emergency funding.
By 2008, bank capital levels had sunk to an all-time low; bank managers and their regulators, believing that risk could be perfectly monitored and controlled, were comfortable with the trend. By 2007, banks’ “leverage ratio”—the percentage of quality capital relative to their assets—was just 6 percent, well below the nearly 8 percent of a decade earlier. Since then, thanks to tougher rules, the leverage ratio has risen above 9 percent. Global capital ratios have risen, as well. Many analysts believe that capital requirements should be higher still, but the shift has made banks somewhat safer.
The government doesn’t mandate capital levels with the goal of keeping any particular bank safe. After all, private companies go out of business all the time, and investors in any private venture should be prepared to take that risk. The capital requirements are about keeping the economy safe. Banks tend to hold similar assets—various types of loans to people, businesses, or government. So when one bank gets into trouble, chances are that many others are suffering as well. A higher capital reserve lessens the chance of several banks veering toward insolvency simultaneously, which would drain the economy of credit. It was that threat—an abrupt shutdown of markets for all lending, to good borrowers and bad—that led Washington to bail out the financial industry (mostly the banks) in 2008.
But what if the financial industry, in creating credit, bypasses the banks? According to the global central banks and regulators who make up the international Financial Stability Board, this type of lending constitutes “shadow banking.” That’s an imprecise, overly ominous term, evoking Mafia dons writing loans to gamblers on betting slips and then kneecapping debtors who don’t pay the money back on time, but the practice is nothing so Tony Soprano-ish. The accountancy and consultancy firm Deloitte defines shadow banking, wonkily, as “a market-funded credit intermediation system involving maturity and/or liquidity transformation through securitization and secured-funding mechanisms. It exists at least partly outside of the traditional banking system and does not have government guarantees in the form of insurance or access to the central bank.”
In plain English, “maturity and/or liquidity transformation” is exactly what a bank does: making a long-term loan, such as a mortgage, but funding it with short-term deposits or short-term bonds. Outside of a bank, the activity involves taking a mortgage or other kind of longer-term loan, bundling it with other loans, and selling it to investors—including pension funds, insurers, or corporations with large amounts of idle cash, like Apple—as securities that mature far more quickly than the loans they contain. The risks here are the same as at the banks, but with a twist: if people and companies can’t pay off the loans on the schedule that the lenders anticipated, all the investors risk losing money. Unlike small depositors at banks, shadow banks don’t have recourse to government deposit insurance. Nor can shadow-financing participants go to the Federal Reserve for emergency funding during a crisis—though, in many cases, they wouldn’t have to: pensioners and insurance policyholders generally don’t have the right to remove their money from pension funds and insurers overnight, as many bank investors do.
Understood broadly, shadow banking is nothing new, encompassing everything from corporate bond markets to payday lending. And much of it isn’t very shadowy; as a recent U.S. Treasury report noted, the government “prefers to transition to a different term, ‘market-based finance,’ ” because applying the term “shadow banking” to entities like insurance companies could “imply insufficient regulatory oversight,” when some such sectors (though not all) are highly regulated. It isn’t always easy to separate real banks from shadow banks, moreover. Just as before the financial crisis, banks continue to offer shadow investments, such as mortgage-backed securities or bundled corporate loans, and, conversely, banks also lend money to private-equity funds and other shadow lenders, so that they, in turn, can lend to companies.
Such market-based finance has its merits; sound reasons exist for why a pension-fund administrator doesn’t just deposit tens of billions of dollars at the bank, withdrawing the money over time to meet retirees’ needs. For people and institutions willing, and able, to take on more risk, market-based finance can offer higher interest rates—an especially important consideration when the government keeps official interest rates close to zero, as it did from 2008 to 2016. Shadow finance also offers competition for companies, people, and governments unable to borrow from banks cheaply, or whose needs—say, a multi-hundred-billion-dollar bond to buy another company—would be beyond the prudent coverage capacity of a single bank or even a group of banks.
Theoretically, bond markets and other market-based finance instruments make the financial system safer by diversifying risk. A bank holding a large concentration of loans to one company faces a major default risk. Dispersing that risk to dozens or hundreds of buyers in the global marketplace means—again, in theory—that in a default, lots of people and institutions will suffer a little pain, rather than one bank suffering a lot of pain.
But too much of a good thing is sometimes not so good, and, in this case, the extension of shadow banking threatens to reintroduce the risks that innovation was supposed to reduce. Recent growth in shadow banking isn’t serving to disperse risk or to tailor innovative products to meet borrowers’ needs. Two less promising reasons explain its expansion. One is to enable borrowers and lenders to skirt the rules—capital cushions—that constrain lending at banks. The other—after a decade of record-low, near-zero interest rates as Federal Reserve policy—is to allow borrowers and lenders to find investments that pay higher returns.
The world of market-based finance has indeed grown. Between 2002 and 2007, the eve of the financial crisis, the world’s nonbank financial assets increased from $30 trillion to $60 trillion, or 124 percent of GDP. Now these assets, at $160 trillion, constitute 148 percent of GDP. Back then, such assets made up about a quarter of the world’s financial assets; today, they account for nearly half (48 percent), reports the Financial Stability Board (FSB).
Within this pool of nonbank assets, the FSB has devised a “narrower” measure of shadow banking that identifies the types of companies likely to pose the most systemic risk to the economy—those most susceptible, that is, to sudden, bank-like liquidity or solvency panics. The FSB believes that pension funds and insurance companies could largely withstand short-term market downturns, so it doesn’t include them in this riskier category. That leaves $45 trillion in narrow shadow institutions and investments, a full 72 percent of it held in instruments “with features that make them susceptible to runs.” That’s up from $28 trillion in 2010—or from 66 percent to 73 percent of GDP.
Of that $45 trillion market, the U.S. has the largest portion: $14 trillion. (Though, as the FSB explains, separation by jurisdiction may be misleading; Chinese investment vehicles, for example, have sold hundreds of billions of dollars in credit products to local investors to spend on property abroad, affecting Western asset prices.) Compared with this $14 trillion figure, American commercial banks’ assets are worth just shy of $17 trillion, up from about $12 trillion right before the financial crisis. Banks as well as nonbank lenders have grown, in other words, but the banks have done so under far stricter oversight.
An analysis of one particular area of shadow financing shows the potential for a new type of chaos. A decade ago, an “exchange-traded fund,” or ETF, was mostly a vehicle to help people and institutions invest in stocks. An investor wanting to invest in a stock portfolio but without enough resources to buy, say, 100 shares apiece in several different companies, could purchase shares in an ETF that made such investments. These stock-backed ETFs carried risk, of course: if the stock market went down, the value of the ETF tracking the stocks would go down, too. But an investor likely could sell the fund quickly; the ETF was liquid because the underlying stocks were liquid.
Over the past decade, though, a new creature has emerged: bond-based ETFs. A bond ETF works the same way as a stock ETF: an investor interested in purchasing debt securities but without the financial resources to buy individual bonds—usually requiring several thousand dollars of outlay at once—can purchase shares in a fund that invests in these bonds. Since 2005, bond ETFs have grown from negligible to a market just shy of $800 billion—nearly 10 percent of the value of the U.S. corporate bond market.
These bond ETFs are riskier, in at least one way, than stock ETFs. Some bond ETFs, of course, invest solely in high-quality federal, municipal, and corporate debt—bonds highly unlikely to default in droves. Default, though, isn’t the only risk: suddenly higher global interest rates could cause bond funds to lose value (as new bonds, with the higher interest rates, would be more attractive). And with the exception of federal-government debt, even the highest-quality bonds aren’t as liquid as stocks; they have maturities ranging anywhere from hours remaining to 100 years.
Investors in bond-based ETFs, then, face a much bigger “liquidity” and “maturity” mismatch risk. If the investors want to sell their ETF shares in a hurry, the fund managers might not be able to sell the underlying bonds quickly to repay them, particularly in a tense market. That’s especially true, since bond markets are even less liquid than they were pre–financial crisis. Because of new regulations on “market making,” banks will be highly unlikely to buy bonds in a declining market to make a buck later, after the panic subsides.
A look at a related type of debt-based ETF raises even bigger mismatch concerns. “In 2017, investors poured $11.5 billion into U.S. mutual funds and exchange-traded funds that invest in high-yield bank loans,” notes Douglas J. Peebles, chief investment officer of fixed-income—bonds—at the AllianceBernstein investment outfit. A high-yield bank loan is one that carries particular risk, such as a loan to a company with a poor credit rating or to a company borrowing money to merge with another firm or to expand; the “yield” refers to the higher interest rate required to compensate for this risk. Rather than keep this loan on its books, the bank is selling it, in these cases, to the exchange-traded funds that are a rising component of shadow banking.
This new demand has induced lending that otherwise wouldn’t exist—in many cases, for good reason. “The quality of today’s bank loans has declined,” Peebles observes, because “strong demand has been promoting lax lending and sketchy supply. . . . Companies know that high demand means they can borrow at favorable rates.” Further, says Peebles, “first-time, lower-rated issuers”—companies without a good track record of repaying debt—are responsible for the recent boom in loan borrowers, from fewer than 300 institutions in 2007 to closer to 900 today. The number of bank-loan ETFs (and similar “open-ended” funds) expanded from just two in 1992 to 250 in June 2018.
Peebles worries as well about the extra risk that this financing mechanism poses to investors. “In the past, banks viewed the loans as investments that would stay on their balance sheets,” he explains, but now that banks sell them to ETFs, “most investors today own high-yield bank loans through mutual funds or ETFs, highly liquid instruments. . . . But the underlying bank loan market is less liquid than the high-yield bond market,” with trades “tak[ing] weeks to settle.” He warns: “When the tide turns, strategies like these are bound to run into trouble.”
The peril to the economy isn’t just that current investors could lose money in a crisis, though big drops in asset markets typically lead people to curtail consumer spending, deepening a recession. The bigger danger is a repeat of 2008: fear of losses on existing investments might lead shadow-market lenders to cut off credit to all potential new borrowers, even worthy ones. Banks, because they’re dependent on shadow banks to buy their loans, would be unlikely to fill the vacuum. “Although non-bank credit can act as a substitute for bank credit when banks curtail the extension of credit, non-bank and bank credit can also move in lockstep, potentially amplifying credit booms and busts,” says the FSB. The porous borders between the supposedly riskier parts of the nonbank financial markets—ETFs—and the less risky ones also could work against a fast recovery in a crisis. Thanks to recent regulatory changes, insurance companies, for example, are set to become big purchasers of bond ETF shares.
Worsening this hazard, just as with the collateralized debt securities of the financial meltdown, many bond-based ETFs contain similar securities. Such duplication could eradicate the diversification benefit that the economy supposedly gets from dispersing risk. Contagion would be accelerated by the fact that debt-based ETFs, like stock-based ETFs, must “price” themselves continuously during the day, according to perceived future losses; this, in effect, introduces the risk of stock-market-style volatility into long-term bond markets. (Bond-based mutual funds, of course, have existed for decades, but they did not trade like stocks and thus did not feature this particular risk.) Via the plunging price of collateralized debt obligations, we saw, in 2008, what happened to the availability of long-term credit when exposed to the pricing signals of an equity-style crash, but those collateralized debt obligations traded far less frequently than bond ETFs do today. Bond ETFs may be more efficient, yes, in reflecting any given day’s value; that supposed benefit could also allow a panic to spread more rapidly.
During the last global panic, the answer to getting credit flowing again—so that companies could perform critical tasks, such as meeting payrolls, before revenue from sales came in—was to provide extraordinary government support to the large banks. But even if one believes that such bailouts are a sensible approach to financial crises—a highly tenuous position—how would the government provide longer-term support to hundreds of individual funds, to ensure that the broader market keeps functioning for credit-card and longer-term corporate debt? This would greatly expand the government safety net over supposedly risk-embracing financial markets—by even more than it was expanded a decade ago.
Unwise lending also harms borrowers. Private-equity firms, too, are increasingly lending companies money, instead of just buying those firms outright, their older model. As the Financial Times recently reported, private-equity funds—or, more accurately, their related private-credit funds—have more than $150 billion in money available for investment. They make loans that banks won’t, or can’t, make, though this is leading banks to take greater risks to compete. “It’s been great for borrowers,” says Richard Farley, chair of law firm Kramer Levin’s leveraged-finance group, as “there are deals that would not be financed,” or would not be financed on such favorable terms.
Competition is usually healthy, and risky finance can spark innovation that otherwise wouldn’t have happened. But easy lending can also make economic cycles more violent. Even in boom years, excess debt can plunge firms that otherwise might muddle through a recession deep into crisis, or even cause them to fail, adding to layoffs and consumer-spending cutbacks. We can see this happening already, as the Financial Times reports, with bankrupt firms like Charming Charlie, an accessories store that expanded too fast; Six Month Smiles, an orthodontic concern; and Southern Technical Institute, a for-profit technical college.
The numbers are troubling. The expansion of shadow banking has unquestionably brought a pileup of debt. The Securities Industry and Financial Markets Association, a trade group, estimates that U.S. bond markets, overall, have swollen from $31 trillion to nearly $42 trillion since 2008. Federal government borrowing accounts for a lot of that, but not close to all of it. The corporate-bond market, for example, went from $5.5 trillion to $9.1 trillion over the same decade. Corporations, in other words, owe almost twice as much today in bond obligations as they did a decade ago. That’s sure to make it harder for some, at least, to recover from any future downturn.
There are policy approaches to resolving these debt issues. An unpopular idea would be to treat markets that act like banks, as banks—requiring ETFs, say, to hold the same capital cushions and adhere to the same prudence standards as banks. In the end, though, the bigger problem is cultural and political. What we’re seeing, more than a decade after the financial crisis, results from the government’s mixed signals about financial markets. On the one hand, the U.S. government, along with its global counterparts, realized in 2008 that debt had reached unsustainable levels; that’s partly why it sharply raised bank capital requirements. On the other hand, the government recognized that the economy is critically dependent on debt. Absent large increases in workers’ pay, consumer and corporate debt slowdowns would stall the economy’s until-recently modest growth. That’s why the U.S. and other Western governments have kept interest rates so low, for so long.
Thus, we find ourselves with safer banks but scarier shadows. Global debt levels are now $247 trillion, or 318 percent, of world GDP, according to the Institute of International Finance, up from $142 trillion owed in 2007, or 269 percent of GDP. When both regular banks and shadow banks are tapped out, we may have to invent shadow-shadow finance to take up the slack.
Top: Illustration by The Heads of State