Systemic Risk: Definition and Application

Market valuations are reaching all-time highs. Liquidity in the bond market has been adversely affected by recent European Central Bank (ECB) policy changes, while low interest rates are encouraging many investors to purchase high-yield, high-risk assets such as junk bonds, leveraged loans, and covenant-lite loans. Central banks around the world continue to pump massive liquidity into the bond markets, while yields have shrunk so much that roughly 30% of the global bond market is delivering negative interest rates.

Since the crisis of 2008, the United States has further lowered lending standards as government-sponsored enterprises (GSEs) reduced required down payments from 5% to 3%, and the Federal Housing Finance Agency (FHA) slashed insurance premiums. Meanwhile, according to the US Congressional Budget Office (CBO), US federal spending is growing substantially faster than economic growth.

With all of these obvious problems, it’s hard to believe systemic risk is low. Maybe it is. But then again, maybe it isn’t. How do we know?

When I reviewed the literature to see how the academic and professional financial community defined systemic risk, I discovered an abundance of definitions. A paper disseminated in 2012 by the Office of Financial Research (OFR) studied 31 quantitative measures of systemic risk in the economics and finance literature, providing a nice overview of each at that point in time.

Since 2008, there has been a proliferation of white papers, empirical studies, and financial models. To my surprise, many of these models are largely incompatible with one another, while some are incomplete, addressing only portions of the financial system. Still others are so rigidly defined that should conditions change, the models would become obsolete. Some are too specific and fail to capture dynamic changes in the structure of the financial system: In trying to solve one problem, they introduce several more. Others focus almost entirely on empirical analysis of certain security prices, while some are purely mathematical.

To be sure, systemic risk is a large, complex subject, and many of these authors deserve credit for their insight and observations. Nevertheless, I came away from the process with a sense of disappointment. How can the everyday investor judge systemic risk? Is it high today? Is it low? Are we required to master some 30 or more models based on fundamentally different principles and hope that we get it right?

Even the most cursory study reveals that systemic risk means different things to different people, so I worked to create a definition that potentially everyone could agree on. Given the widely divergent views, a working definition needs to balance specificity in targeting the sources and nature of systemic risk and yet be flexible enough to capture dynamic changes in the structure of the financial system over time.

For instance, when a country lowers its lending standards, many loans could be classified as “prime” when they should really be “sub-prime.” A model myopically focused on the amount of sub-prime mortgages, for instance, might miss this critical change. In finance (as well as in politics), there are often multiple ways to achieve the same end, so a definition that is either too specific or too inflexible inevitably leads to measurement error, creating new opportunities for very old sins.

So, what is the definition of systemic risk?

“Systemic risk is the risk of a large-scale failure of a financial system whereby a crisis occurs when providers of capital (depositors, investors, and capital markets) lose trust in either the users of capital (banks, borrowers, leveraged investors, etc.), or in a given medium of exchange (the US dollar, Japanese yen, pound sterling, gold, etc.). Perhaps the most important feature of systemic risk is that the risk spreads from unhealthy institutions to otherwise healthy institutions through a transmission mechanism. (If it were not for this transmission channel, then the risk would not be systemic, but rather fundamental in nature.) This means of transmission is characterized by leverage, interdependence, and buffers throughout the financial system.”

Looking to the United States as an example, let’s examine each component.

Leverage: Debt, Guarantees, Credit Quality, Duration Mismatch, etc.

There are clearly some countervailing forces at work here. On the consumer side, balance sheet repair continues as households delever. In the corporate and government sectors, leverage is now at or near all-time highs. The securities markets remain highly levered, which amplifies market moves considerably — in either direction. Monetary policy has pushed many global investors into low-quality assets which will also amplify leverage in the next down market.

Offsetting these issues are insurance companies that have substantially ramped up their direct lending. Because insurers typically have long duration liabilities (20-year term life policies, for example), they are more natural lenders and experience less duration and refunding risk. In a default cycle, insurers present much less risk to the system than banks, which must refinance their loans as short duration funding expires. From the standpoint of systemic risk, this is a positive development.

Lenders: Banks, Shadow Banks, etc.

Commercial bank leverage has improved over the past several years, but many of the same low-credit policies remain in place.

• Total equity to total assets is at 11.26 for the US banking industry.
• More than 5% of US banks are loosening credit standards on consumer loans.
• In the first quarter, 13% of banks tightened corporate lending standards.

Financial Institutions: Investment Banks, Asset Managers, Pension Funds, etc.

Investment banks and asset managers are naturally very different entities.

• Wall Street is ramping up “debt-on-debt-on-debt.”
• Wall Street is taking on more risk to help companies sell stock.

Corporate

• Non-financial corporate debt is now 45.3% of GDP, approaching all-time highs.
• The median debt/EBITDA ratio is now 2.3, its highest mark since 2000.

Consumer

• Household debt to permanent disposable income (PDI) is down 35% from its peak in 2008.

Government

• The US federal debt/GDP is now 105.1% vs 61.3% just 10 years ago.

Markets: Margin Debt, Margin Requirements, etc.

• Cash-to-margin debt ratios are near an all-time low.

Insurers

• Insurers now account for 11.6% of the global private debt market.

Central Banks

• The US Federal Reserve holds steady at $4.4 trillion total assets.

Super-Regional Organizations: IMF, World Bank, etc.

• Bank of Canada says the IMF’s credit risk has increased markedly.

Interdependence

As far as interdependence goes, many factors are in play. Too-big-to-fail banks have become larger. Fed policy has pushed investors of all types, including banks, to take on riskier assets. More and more borrowers are relying on central bank funding, while bond market liquidity is at an all-time low.

Cross Holding of (Common) Securities, Options, Swaps, etc.

• Research shows that banks create systemic risk for insurers, but not vice versa.

Guarantees

• Use of letters of credit is declining.

Common Funding Sources

Duration Mismatch

• “Goldman: The Search for Yield Is Just Getting Started“

Regulations and Policies That Channel Capital into the Ownership of Specific Securities: Basel Rules, etc.

Liquidity in Markets

• Liquidity in the developed world is now worse than in emerging markets.

Buffers

The world economy is very sluggish, and the ability of the United States to raise revenue by raising tax rates is curtailed at present. With $20 trillion in federal government debt, raising capital via government bond auction may become incrementally more difficult — particularly in the face of the next recession.  Of course, the Fed can continue printing money, but its capacity to offset increased printing with increased debt will diminish year by year unless the trajectory is changed.

The crisis of 2008 has apparently done little to materially improve credit standards. Banks are steadily reducing credit standards again.  In the commercial and industrial sectors, leverage has increased to all-time highs, making it very unlikely that commercial credit growth can continue. In fact, corporate debt leverage is now at record highs, which is very unusual in a growing economy. Subprime auto loans, student loans, and consumer finance are ubiquitous. More recently, subprime mortgages have returned in force. This wave of subprime borrowing has been offset to a significant degree by the deleveraging of the household sector since the crisis.

In the banking sector, loan loss provisions are running near all-time lows, and covenant restrictions are almost absent. Moreover, collateral requirements have been subject to unyielding political pressure and banks are beginning to acquiesce.

Currency Buffers: Foreign Exchange Holdings, Foreign Ownership of Sovereign Bonds, Money Printing, Taxation Ability, Capacity to Set Interest Rates, Current Account Balances, etc.

• “Negative Rates Push Japanese Investors into US Treasuries“
• “As China Dumps Treasuries, World Sees No Better Place for Refuge“

Down Payments: Mortgage Down Payments, Collateralized Loans, etc.

• Subprime mortgage origination was up 38.7% in May 2016 over May 2015.

Capital Provider Buffers: Loan Loss Reserves, Loan Covenants, Over-Collateralization of Asset-Backed Securities, Margin Capital, etc.

• Bank loan loss provisions are running near a 30-year low.
• Covenant-lite loans are at all-time highs.

Direction of Buffers

• Delinquency rates are rising.
• Fannie Mae reduces the mandatory waiting period after bankruptcy.

One risk of the extraordinary monetary response to the financial crisis of 2008 was material inflation. But the inflationary force of enormous money printing — quantitative easing (QE) — was offset by the weakness in the real economy and the assumption of substantial debts by the federal government and the Fed since issuing sovereign debt helps protect the value of a currency. However, the financial flexibility of the government and the Fed is greatly reduced today compared to what it was before the crisis. Systemic risk has shifted into the corporate and government sectors.

With respect to the US government, spending on existing social programs will continue to exceed GDP growth. If there are no changes to existing programs, the federal budget will experience substantial pressure regardless of the economic environment. But in a recession, the pressure will be amplified. So, the next recession means one of two things: increased solvency risk or increased inflation risk. In the case of the United States, the dollar still maintains its status as global reserve currency and the Fed can continue to print money, so solvency is not at stake. However, country’s ability to issue debt at low rates could be tested.

The banks will experience turbulence with defaulting corporates, but exposures to corporates are probably not significant enough to cause contagion. So, the real pressure — the systemic risk — likely lies with the Fed as a common funding source and US Treasury bonds as a common issue. In turn, substantial owners of Treasury bonds will likely risk taking large marks on this section of their balance sheets. This is where any contagion will begin.

Organizing information in this way paints a clearer picture of what systemic risk is and where it may be going. The next crisis won’t be a mere repeat of the last one. It will reflect where our policies, investments, and actions have taken us — leaving a trail back right to the original source.

A new economic database, Financial Scandals, Scoundrels, & Crises, is now online. It features in-depth analysis of the 2008 financial crisis, 1998 Russian default, and the failure of Long-Term Capital Management (LTCM). Coverage of additional topics is forthcoming.

If you liked this post, don’t forget to subscribe to the Enterprising Investor.

All posts are the opinion of the author. As such, they should not be construed as investment advice, nor do the opinions expressed necessarily reflect the views of CFA Institute or the author’s employer.

Image credit: ©iStockphoto.com/erhui1979