Credit Crises: The Role of Excess Capital

By Bruce G. Stevenson

Credit crises are catastrophic events in which banks and lenders suffer extreme losses when loans and other credit instruments default on a large scale and cause banks to fail in extraordinary numbers. Massive loss of economic value ensues, threatening the viability of national economies and the global financial system.

The most recent credit crises, the 2007 Subprime Mortgage Crisis and 2007-2009 Great Recession, have striking parallels to the Roaring Twenties and the Great Depression. In both periods, rapid increases in the value of residential real estate fueled speculation in the housing and equity markets, and when the real estate bubbles burst, massive recessions and unemployment followed. In the eighty years between these catastrophes, several other credit crises occurred including a real estate investment trust crisis in the mid-1970s and a commercial real estate crisis in the late 1980s and early 1990s.

Credit Crises: The Role of Excess Capital provides the first definitive explanation for these repetitive catastrophes: the Excess Capital Hypothesis (ECH). Written for bankers, bank regulators, finance professionals, and policymakers, Credit Crises provides a detailed explanation of how excess capital has been the driver of past credit crises. The ECH is the definitive roadmap for mitigating credit crises, and Credit Crises offers recommendations to bankers, bank regulators, and policymakers on how to prevent and lessen future crises.

• Language: English
• Publisher: Gatekeeper Press
• Release date: Mar 7, 2024
• ISBN: 9781662941986

Book preview

Introduction

Since the global financial crisis and recession of 2007-2009, criticism of the economics profession has intensified. The failure of all but a few professional economists to forecast the episode – the aftereffects of which still linger – has led many to question whether the economics profession contributes anything significant to society.

Robert J. Schiller, Sterling Professor of Economics, Yale University and Fellow, International Center for Finance, Yale University School of Management, 2015

One intriguing subplot of the economic crisis is the failure of most economists to predict it. Here we have the most spectacular economic and financial crisis – possibly since the Great Depression – and the one group that spends most of its waking hours analyzing the economy basically missed it.

Robert J. Samuelson, Journalist, The Washington Post, 2009

The collapse of the U.S. residential mortgage market in 2007 and 2008 precipitated the worst market contraction in nearly 100 years, the so-called Financial Crisis, and created the Great Recession, the most significant economic recession since the Great Depression. During the Financial Crisis, several U.S. investment banks and many monoline mortgage companies failed, causing government officials to be concerned about the safety of the U.S. banking system. Policymakers and banking regulators formulated several responses, including forcing liquidity into the banking system to ensure that more banks did not fail.

During the Financial Crisis and Great Recession, all investors in risky assets² suffered as prices for those assets declined dramatically. For example, national home prices contracted by more than one-third, based upon the decline of the Case-Schiller Home Price Index from 2006 to 2009, and by more in various regions across the country, including Florida and the Northeast. The Dow Jones Industrial Average fell by 54 percent between October 2007 and March 2009, and the S&P 500 fell by 57 percent in the same period. Gross Domestic Product (GDP) in the United States declined by 0.3 percent in 2008, and 2.8 percent in 2009, and unemployment briefly reached 10 percent in 2009.

In my judgment, it is unfair to paint the entire economics profession as having failed to predict the collapse of the mortgage market and the Great Recession. People are psychological creatures, and those around them shape their thoughts and ideas. In the period leading up to the Financial Crisis, there was ebullience—irrational exuberance, if you will. It would have been tough to be a lone voice saying that what appeared to be good, sound, and rational behavior, such as believing that there was no bubble in the housing market because housing prices couldn’t and wouldn’t fall, was none of those things. Consequently, very few professionals predicted these crises, including the bankers and financiers who contributed so much to creating them.

The U.S. financial system and the global economy have recovered from the Financial Crisis and Great Recession, yet the root causes of these events remain in debate. We still don’t have a firm understanding of why credit crises occur, even though they are relatively common and very costly. Considering the tremendous economic and social consequences of the most extreme credit crises, such as those leading up to the Great Depression and the Great Recession, it is surprising that they have received relatively little attention from academics and experts in the financial services industry.

1.1 Bank Failures Are Episodic

This situation is striking, given that these crises periodically put the banking system of the United States at significant risk. For example, Figure 1-1 shows the pattern of bank failures from 1934 to 2020, and it is clear that three crises define this pattern: the collapse of the stock market in 1929 followed by the Great Depression; the collapse of commercial real estate markets at the end of the 1980s and the beginning of the 1990s; and the subprime mortgage crisis and Great Recession in the late 2000s. These are the worst credit crises since 1900.

These waves of bank failures effectively wiped out the investments made by holders of bank stocks. This loss of investment capital is important, of course, to the individual investors, but it is also crucial to the financial markets as a whole. The role of banks as financial intermediaries is critically important to all financial markets: within local communities, in the global financial markets, and everything in between. Without equity investors in banks, there would be no banks and no capital to support deposit accounts for both individuals and corporations, as well as loans to individuals and corporations. Investors in bank stocks must expect a market-level rate of return to make these investments in the first place, and bank failures prevent them from realizing those rates of return.

Figure 1-1 Failed Banks in the U.S Commercial Banking Industry, 1934 – 2021

(Source: Federal Deposit Insurance Corporation)

In the modern era, the Federal Deposit Insurance Corporation (FDIC) insures deposit accounts at banks and thrifts in the United States. When a bank fails, the FDIC reimburses depositors up to the insured limit for a given category of deposit account at the failed institution. While no depositor has lost any money in insured accounts since FDIC insurance began in 1934, extreme credit crises do threaten the insurance fund if large numbers of banks, very large banks, or both, fail. The financial markets and depositors both count on the solvency of the FDIC insurance fund.

Under normal circumstances, a depositor must place deposits at more than one bank to receive deposit insurance above the FDIC limit at any single bank. However, in rare instances, the FDIC will insure deposits above the insured limit at a given bank. In so doing, however, it risks creating moral hazard. Moral hazard is the lack of an incentive to manage risk when that risk is effectively transferred to another party, as is the case when the FDIC bears the losses of a depositor.

For these reasons, and others, it is vital to know the typical characteristics of credit crises, why they take place, and how the supervision of banks can be enhanced to prevent such crises in the future or to ameliorate their consequences.

Throughout my career, I have been interested in why credit crises occur and why they occur with seeming regularity. This interest grew out of the experiences I had in living and working through three market crises: the meltdown of commercial real estate (CRE) markets in the late 1980s and early 1990s, the Asian financial crisis in the late 1990s, and the Financial Crisis and Great Recession of the late 2000s. Even earlier, there was the failure of real estate in the mid-1970s that led to the collapse of the real estate investment trust (REIT) market. Moreover, of course, the Great Depression and its real estate crisis of the 1920s and 1930s defined the U.S. economy and the banking sector for decades, even after World War II.

I began my career in banking at Citigroup in 1987 as that bank was building a portfolio of CRE loans that would ultimately become one-quarter of all loans it made to individual and commercial borrowers in North America. I had a ring-side seat to the collapse of the CRE markets and the near collapse of Citibank.

Later, I led the portfolio management division of BNP Paribas when the Asian Financial Crisis struck, during which the large and visible hedge fund, Long Term Capital Management (LTCM), failed. BNP Paribas worked with many other banks, under the guidance of the Federal Reserve, to provide liquidity so that the resolution of LTCM occurred gradually. A few years later, I managed the impact of the Dot-Com crisis on BNP Paribas’s commercial loan portfolio, including those internet and media assets we had securitized in a collateralized loan obligation (see Chapter Fourteen).

Finally, I joined HSBC in August 2006, too late to influence that bank’s purchase of Household Finance and Household’s portfolio of subprime residential mortgages, but in time to watch that portfolio implode, and losses grow. Furthermore, we spent the better part of three years estimating and re-estimating how significant the mortgage losses would be and how much capital our parent company, HSBC Ltd., would have to put into its North American banking subsidiary to keep it afloat. Shortly after the mortgage market collapsed, it became clear that Household Finance would have failed had it been a stand-alone company in 2008. Instead, the shareholders of HSBC Ltd. bore the brunt of this multi-billion-dollar fiasco, now widely recognized as a mistake for HSBC and the markets, since the purchase of Household lent justification for the wild excesses that characterized mortgage lending in the mid-2000s.

Some of the most intelligent people I have ever known were my colleagues in commercial banks. However, most failed to foresee any of these crises, and almost none forecast all three. One exception is my friend Michael Fadil, with whom I worked at Fleet Financial Group in the mid-1990s. He and I spent many hours building tools to assist our colleagues in Fleet’s strategic planning unit value distressed banks and distressed loan portfolios. Fleet was growing rapidly by purchasing banks that were suffering from the CRE crisis of the late 1980s and early 1990s.

Michael and I often discussed why banks routinely got themselves into trouble, not fully understanding why these institutions, whose job was to understand, price, and manage credit risk, seemingly could do none of the three in credit crises.

1.2 Default Rates Follow an Exponential Pattern

Those conversations took place shortly after the major rating agencies began publishing studies on the default experience of corporate bond issuers in the U.S. These studies confirmed and quantified what seasoned participants in the bond market already knew from experience: default rates were geometrically greater for weaker borrowers than for stronger borrowers. Put differently, default rates increase exponentially from AAA to CCC, as shown in Figure 1-2.

Figure 1-2 Weighted Long-Term Average Default Rates in the US. Corporate Bond Market, 1981 to 2021

(Source: Kraemer & Gunter, 2022)

With the publication of these studies, the distinction between investment grade³ bonds and non-investment grade bonds became starkly apparent. For issuers with investment grade ratings or those ratings between AAA and BBB, default rates are de minimis on an annual basis. Annual default rates become material only when the issuer’s rating is non-investment grade (BB through C).

Subsequently, evidence emerged that this pattern of exponential increase in default rates with declining credit quality is present in many classes of debt, including municipal and consumer debt (e.g., VantageScore, 2011)⁴. Similar to the distinction of investment grade and non-investment grade for corporate borrowers, consumer borrowers are classified according to their credit quality, with prime applied to the best quality borrowers and subprime applied to consumers with higher credit risk. Typically, these categories are defined based on a borrower’s credit scores, and default or delinquency rates increase as credit scores worsen from prime to subprime to deep subprime.

Figure 1-3 shows one example published by the Dallas Federal Reserve Bank in which the relationship between consumer credit score and the rate of serious delinquency (the percentage of the loan balance that is at least 90 days past due or written off as a loss or foreclosure) for consumers in the Dallas, Texas metropolitan region (Perlmeter & Groves, 2018).

Danis and Pennington-Cross (2008) examined a large sample of securitized private-label residential mortgages and found that borrower credit scores were strong predictors of delinquency and default on those loans.

Figure 1-3 Relationship of FICO Scores to Serious Delinquencies on Consumer Loans, Federal Reserve Report for Dallas, Texas

(Source: Perlmeter & Groves, 2018)

1.3 Credit Crises Are Related to Default Rates of Borrowers

Michael and I concluded that there must be a relationship between these exponential patterns of delinquency and default and the occurrence of credit crises. In 1994, we published an article presenting the idea that credit crises occur due to the credit risk characteristics of banking customers (Stevenson & Fadil, 1994). We discovered that credit crises could emerge as the simple mathematical result of long-term default rates. The more credit banks extend to non-investment grade borrowers, the more likely those borrowers are to default and default en masse.

For example, according to our Monte Carlo simulations, a portfolio of BB-rated borrowers will experience a default rate of at least 5 percent once every 11.5 years. That same portfolio will experience a default rate of at least 8 percent once every 23.3 years. For a portfolio of B-rated borrowers, defaults occur much more frequently. That B-rated portfolio has a 65.3 percent chance of a default rate of 5 percent each year and a 34.8 percent chance of an 8 percent default rate each year.

However, our paper did not answer these core questions:

● Can it be the case that credit crises simply are the mathematical product of the default probabilities of underlying borrowers?

● If so, why do banks spend so much effort and money to understand and manage credit risk?

● If so, why is there such a big burden of regulation on banks and the capital markets?

● If there are more reasons for credit crises beyond simple probabilities of default, what are those reasons?

● If it is the business of banks to lend to creditworthy customers (i.e., those with low or de minimis default probabilities), why do credit crises occur, and why do they occur with seeming regularity?

● Why do defaults and losses occur in waves, as in credit crises, rather than randomly and without a pattern?

Bank failures seem to follow a pattern broadly similar to the hypothetical simulations that Michael and I conducted since actual bank failures occur not randomly over time but in waves. For example, the peak years of bank failures since 1970 are separated by 12.5 years, on average (see Figure 1-1). That bank failures occur in waves, with failures clustered in short periods, gives rise to the concept of crisis.

Later, we will see that delinquencies, defaults, and losses on a broad array of loans follow these same patterns. The similarity suggests an underlying cause or series of causes rather than just the consequence of mathematical probabilities.

This book examines these questions and proposes answers to them. I argue that credit crises, defined as eruptions in defaults on credit agreements that result in extreme losses to lenders and investors and sometimes to bank failures, result from excessive lending in discrete periods or waves. Excessive lending drives capital to increasingly risky borrowers, who default and generate losses. In response, lenders become risk-averse, cease to lend to risky customers, withdraw loans from the credit markets, and invest their capital in riskless securities. By withdrawing capital from the markets, banks precipitate more defaults. I call this idea the Excess Capital Hypothesis (ECH).

Lending in excess of what? Lending in excess of economic growth. The ECH holds that when excess capital emerges in the market, it does so because the rate of growth in debt exceeds the rate of economic growth. When this happens, debt flows to increasingly risky borrowers (non-investment grade or subprime) whose probabilities of default increase exponentially.

1.4 The Definition and Function of Commercial Banks

At this early stage of our examination of credit crises, it is useful to specify the function of U.S. commercial banks. Each of these elements bears on how credit crises are created and managed and the role of excess capital in credit crises.

I will begin with the definition of a commercial bank. A commercial bank is a financial institution that accepts deposits, offers checking accounts, savings accounts, and basic financial products, such as certificates of deposit, and makes loans to individuals and businesses. Customer deposits provide banks with the capital to make these loans which then become the most essential assets on banks’ balance sheets.

1.4.1 Income and Expenses of Commercial Banks

The income earned by a bank comes from two sources: (1) interest income; that is, the revenues earned from the interest charged by the bank on loans that are paid to the bank by its borrowers, and (2) non-interest income, which is the revenues earned by the bank from sources other than loan interest rates, including fees and revenues earned from trading activities.

The principal expenses of banks also come from two sources: (1) interest expense, which is the expense of interest paid on the deposits, and (2) non-interest expense, which is an operating expense of a bank, including employee salaries, bonuses, and benefits; equipment rental; costs of information technology; rent; taxes; professional services; and marketing.

Figure 1-4 Interest Income and Non-Interest Income for U.S. Commercial Banks, 1970 to 2021

(Source: Federal Deposit Insurance Corporation)

Interest income is the dominant source of revenues for U.S. commercial banks, and, on average, from 1970 to 2021, interest income is five times the size of non-interest income (Figure 1-4). Further, net interest income, the difference of interest income minus interest expense, is a commercial bank’s most important source of net revenues.

1.4.2 Loan Accrual and Non-Accrual

Typically, a borrower is expected to make regular payments of interest and principal on its loan(s) (e.g., monthly payments on a residential mortgage). However, there are some loans for which all but the last payment are payments of loan interest, and the last payment consists of the final interest payment and all of the loan principal.

In its accounting system, a commercial bank recognizes the contractual nature of a loan, including the planned payments of interest and principal. As it records interest income, the bank includes scheduled interest payments before those payments are made. This process is known as accrual, and accrued interest is the amount of interest incurred on a loan or other financial obligation as of a specific date but has not yet been paid. Accrued interest is interest revenue for the bank and it is interest expense for the borrower.

However, loans carry credit risk, that is, the probability that a borrower will cease to make payments of principal and interest on its loan, known as a default. As the bank considers extending a loan to a borrower, it scrutinizes that borrower and its ability to repay the loan. Based on this analysis, the bank sets the loan structure, such as whether the loan is secured by collateral, and determines the borrower’s sources of income for repayment of both the loan principal lent by the bank and the interest charged on the loan. This process is known as underwriting.

When a borrower has not made a payment on a loan for 90 days (i.e., 90 days past due), default on that loan is imminent or has occurred. At the 90-day-past-due stage, the bank ceases to recognize income from loan interest and principal as accruing, and that loan enters a stage known as non-accrual. Non-accrual means that the lender cannot add the interest payment on the loan to its revenue until the payment is made. The accounting for non-accruing loans is cash basis since revenue is recognized on the bank’s income statement only when payments are made and cash is received.

Non-accrual is also known as non-performing and non-accrual status can be applied to all types of loans. Only loans fully secured by highly liquid collateral can be 90 days past due in interest payments and still be considered performing since the sale of the collateral will allow the bank to recover the unpaid loan principal⁵.

1.4.3 Loan Charge-Offs and Recoveries

Non-accruing or non-performing loans generate losses to commercial banks because non-payment of loan principal typically occurs at the same time that non-payment of interest occurs. When a bank determines that it will not receive repayment of the amount it has lent because the borrower is in default, it must recognize that loss of principal in its accounting system. It does so in two ways.

First, when a bank determines that a borrower will default on its loan, it establishes a loan loss reserve on its balance sheet to cover the portion of the loan for which it expects non-payment (a portion of the loan or its entire outstanding balance). In bank accounting, outstanding loans are recorded on the asset side of a bank’s balance sheet. The account for loan loss reserves is a contra-asset account that reduces outstanding loans by the amount the bank’s managers expect to lose when some portion of the loan is not repaid.

Second, the bank’s executives periodically determine how much to add to the loan loss reserves account, and a provision for loan losses is charged against the bank’s current earnings. This loan loss provision is an expense item on the bank’s income statement deducted from revenues before the bank calculates net income. Provisions are made based on losses expected for individual loans in non-accrual and for expected loss on the portfolio of loans not in credit stress (i.e., performing loans) using a pooled actuarial approach.

When these executives determine that all or a portion of the outstanding loan will never be repaid, they charge or deduct from the loan loss reserve account for the lost amount. Specifically, this charge-off is the reduction in the reserve account that a bank believes it will no longer collect once the borrower has defaulted on its payments.

Once a loan secured by collateral defaults, the bank may sell all or a portion of the collateral to recover against the unpaid principal that has been charged off. If the sale proceeds exceed the expected amount, the resulting gain is termed a recovery and is used to offset charge-offs taken by the bank on the loan for which collateral was sold. Net charge-offs on a defaulted loan are charge-offs less recoveries.

1.5 The Definition of a Credit Crisis

A credit crisis is more than just high levels of default on loans or bonds. A credit crisis also is a period in the financial markets when those high levels of defaults trigger extreme losses to lenders, prompting those lenders to tighten standards for lending, withdraw capital from the credit markets, and, in the worst cases, stop lending altogether. As a result, capital is constrained, and markets can experience reduced levels of liquidity. The normal movement of cash in the economy can be disrupted.

Charge-offs on these defaulted loans are losses to banks and adversely impact their capital. During a credit crisis, some banks may become insolvent, fail, and require a takeover by bank regulators (see Figure 1-1). As a result, bank failures tend to be concentrated in the same periods as credit crises, and bank failures may continue even after the crisis eases.

A credit crisis can also be defined mathematically. In this book, a credit crisis occurs when the banking industry experiences charge-offs at or above the 95th percentile of all charge-offs historically⁶. At this level, charge-offs are extremely high and well above the average, or normal, level of losses.

This definition of a credit crisis standardizes the conditions by which historical crises can be evaluated. It permits the comparison of crises in different periods and across different types of assets. For example, the subprime mortgage crisis was the most extreme credit crisis of the modern age in which charge-offs on residential mortgage loans exceeded 1.92 percent of mortgages outstanding from the second quarter of 2009 (2Q2009) to the fourth quarter of 2010 (4Q2010;⁷ see Figure 1-5). The build-up to this crisis in mortgage loans occurred from 3Q2008 to 1Q2009, and the wind-down period took place in 1Q2011 and 2Q2011.

Notably, there were credit crises in other types of consumer loans at this same time, including credit cards (see Figure 1-5). The patterns of losses on mortgages and credit cards were quite similar (Pearson correlation coefficient [r] = 0.708), although the crisis in credit cards was shorter (2Q2009 to 4Q2010).

When I discuss different classes of debt in Section Three, I will refer to this mathematical definition of credit crises (i.e., charge-offs at or above the 95th percentile of historical charge-offs). In Chapter Two, I will explore in more general terms the history of credit crises in the U.S.

Figure 1-5 Charge-Offs on Loans to Consumers, U.S Commercial Banking Industry, 1Q1991 to 2Q2022

(Source: Federal Reserve Bank of St. Louis)

1.6 The Structure of This Book

The book is divided into four sections. Part One examines the history of credit crises in the United States and introduces the ECH as an explanation for the cyclical nature of these crises.

Part Two looks at the role of human psychology and behavior in credit crises, particularly the dynamic nature of credit standards at banks. It also examines similar ideas, including the cyclical patterns of capitalism proposed by Hyman Minsky and the leverage cycle, brought forward by John Geanakoplos of Yale University.

Part Three presents a series of case studies in which I look at different classes of debt and ask whether the ECH can explain the historical pattern of losses on that debt.

Part Four concludes and offers ideas about how investors in debt, especially banks, and their regulators, can use the knowledge provided by the ECH.

Finally, this book is written so that each chapter is a complete story, and while later chapters reference earlier ones, the reader may enjoy each chapter on its own. In other words, feel free to read all of the book from cover to cover or read parts now and read others later.

_________________

² A risky asset is one that holds the potential for loss. Risk can appear as a reduction in the value of the asset or in the reduction in the income produced by the asset. Risk can also appear as volatility in the value of the asset or the income it produces.

³ Investment grade refers to debt with senior unsecured ratings of BBB- or higher. Non-investment grade refers to debt with senior unsecured ratings of BB+ or lower.

⁴ Also see https://www.federalreserve.gov/newsevents/rr-commpublic/cat._1e_meeting_with_fico.pdf and https://www.federalreserve.gov/boarddocs/rptcongress/creditscore/general_tables.htm.

⁵ A loan that is well-secured and in the process of collection can also be 90-days past due and still performing.

⁶ The temporal distribution of charge-offs does not follow a Normal statistical distribution. Rather, that distribution is strongly skewed to the right and the actual 95th percentile occurs at a greater value for charge-offs than would be the case if the distribution were Normal. This skewness implies that, according to this definition of credit crisis, such crises are less frequent and have greater losses than would be the case if the charge-off distribution were Normal.

⁷ The 95th percentile of quarterly charge-offs on residential mortgages from 1Q1990 to 2Q2022 is 1.92 percent. The only period in the string of seven quarters that was below 1.92 percent was 3Q2010 when the rate of charge-offs was 1.90 percent.

PART ONE

Credit Crises and the Excess Capital Hypothesis

A Brief History of Credit Crises in the United States

The interesting thing is that, on the whole, the economics profession didn’t learn the lessons of the East Asian crisis. I wrote about it a great deal, and I continued to do some research on the subject. But because we didn’t learn the lessons of that crisis, we’ve had the crisis that began in 2007.

Joseph E. Stiglitz, Trinity’s Nobel Economists Lecture, 2009

"Why is it possible to rescue S&L buccaneers in the early ‘90s and provide guidance to levered Wall Street investment bankers during the 1998 long term capital management crisis, yet throw 2 million homeowners to the wolves in 2007?"

Bill Gross, Pacific Investment Management Company, 2007

If you remember the early ’90s, other than, I would say, 1928, there was nothing even close. The conditions facing real estate developers in that early ’90s period were almost as bad as the Great Depression of 1929 and far worse than the Great Recession of 2008. Not even close.

Donald Trump, 45th President of the United States and former commercial real estate developer, 2016

The collapse of the global credit markets in 2007 and 2008, which began as a failure of subprime mortgages, was the worst financial crisis since the Great Depression. This catastrophe was very damaging to banks and to the U.S. commercial and investment banking industries. However, such crises are nothing new; the history of the U.S. financial markets is replete with them, and this recent debacle differs only in magnitude.

For the banking industry, the result of these events is self-evident: years of profits can be wiped out in a single crisis. In severe circumstances, individual banks and groups of banks fail (Figure 1-1). Such was the case for the commercial real estate (CRE) crisis of the late 1980s and early 1990s and such was the case for the subprime mortgage debacle of the current era.

This chapter examines the history of credit crises, their implications for the profitability of banking, and their consequences for the commercial banking industry and the financial system as a whole. The central arguments are that such catastrophes are the key to the profitability and viability of commercial lending and that they have common characteristics. Banks that lend sometimes take significant risks, risks that periodically manifest themselves in credit crises and in which the profitability of the bank, and sometimes the bank itself, is threatened.

2.1 Credit Crises and Their Consequences

Crises in the global financial markets, some of which are failures in the loan markets, occur about twice a decade (Figure 2-1). Episodes of illiquidity and market risk can create extreme losses for banks, but credit crises are most often the events that lead to bank failures. For example, the entire U.S. investment banking industry disappeared in the recent subprime mortgage mess, with the failure of Lehman Brothers, the acquisitions of Bear Stearns and Merrill Lynch, and the conversion of Goldman Sachs and Morgan Stanley to commercial bank holding companies.

At the height of the commercial real estate crisis in 1991, the Rhode Island Share and Deposit Indemnity Corporation (RSDIC) failed. RSDIC was a private insurer of deposits in Rhode Island. Its failure emerged due to several factors, particularly the extreme level of losses at banks and other depository institutions that had lent to commercial real estate borrowers. As a result of this failure, the state’s governor closed 45 credit unions, banks, and loan and investment companies (Pulkkinen & Rosengren, 1993).

A credit crisis, or a group of crises, occurred once a decade from 1970 through 2010. In addition to these two examples, I briefly highlight some of the worst credit crises in the modern era. Specifically, Figure 2-1 shows the four major credit crises over the last 50 years.

I note that reliable data on quarterly charge-offs by class of debt cover only the period of 1991 to 2022, so credit crises that occurred before 1991 cannot be evaluated by the 95th percentile method discussed above. They are evaluated judgmentally.

2.1.1 Real Estate Investment Trust Crisis of the Mid-1970s

The 1960s were a decade of rapid economic growth and residential and commercial real estate construction. For example, significant development of lodging facilities occurred as the demand for motels grew alongside the expansion of the interstate highway system. Further, from 1960 to 1970, more than 8,000 shopping centers were built in the United States, more than twice as many as in the 1950s, which was the first decade in which Americans shopped at such centers.

As the 1960s ended, the mood in the U.S. real estate markets was optimistic. Corporate America was not only expanding; it was posting record-high profits. Demand for office buildings was high, and development activity began to surge as capital flowed into these markets. For example, the real estate boom that began in the second quarter of 1967 lasted ten quarters and led to an 11.2 percent increase in real construction. Real estate construction accounted for 3.8 percent of real gross domestic product (GDP) at its peak in this period.⁸

Figure 2-1 Major Market and Credit Crises in the United States Since 1970

Much of the financing of real estate development in this period came from a new vehicle called a real estate investment trust or REIT. Essentially, a REIT is a portfolio of real estate properties structured as a company in which individuals invest capital and receive dividends as their returns. They give investors, especially small investors, access to income-producing real estate. (See the box below.)

The first REITs consisted primarily of mortgages and the industry experienced significant expansion in the late 1960s and early 1970s. The growth primarily resulted from the increased use of mortgage REITs (mREITs) in land development and construction deals. Specifically, the building boom that began in the fourth quarter of 1971 lasted nearly two years and produced a 14.4 percent increase in real construction. Construction accounted for 4.0 percent of real GDP.

By the late 1960s, some major markets were experiencing significantly lower levels of real estate construction, and this was the first such slowdown since the 1920s. However, capital persisted in the CRE markets, and a race for the skies kicked into high gear in the 1970s. By the 1970s, REITs accounted for most of the mortgage capital that financed CRE. Many REITs had borrowed from banks to permit more and larger investments, resulting in many REITs having high financial leverage.

Notably, two significant economic recessions occurred in the first half of the 1970s. During the second recession in the mid-1970s, risky loans and too much financial leverage forced most REITs out of business. An estimated $20 billion to $40 billion in mortgage REIT equity was lost and tarnished the reputation of both types of REITs for more than a decade.

Real Estate Investment Trusts

A Real Estate Investment Trust (REIT) is a company that owns real estate, including commercial properties (such as office buildings, apartment buildings, warehouses, hospitals, shopping centers, hotels, or timberland) or housing (by purchasing mortgages or mortgage-backed securities). A REIT allows individual investors to buy shares in a portfolio of real estate properties, and the investors are compensated from the income earned from the portfolio.

The two main REIT types are equity REITs and mortgage REITs (mREITs). Most REITs operate as equity REITs, allowing investors to invest in portfolios of income-producing real estate. These companies typically own commercial properties leased to tenants, such as office buildings. Equity REITs are required to distribute 90 percent of income as dividends to investors.

Mortgage REITs provide financing for income-producing real estate by purchasing or originating mortgages and mortgage-backed securities (MBS) and earning income from the interest on these investments.

Hybrid REITs use the investment strategies of both equity REITs and mREITs.

REITS can be traded publicly on major exchanges, they can be public but not traded, or they can be private.

Importantly, REITs can use debt as part of their corporate capital structure. REITs tend to use substantial amounts of debt; this leverage enables them to use a fixed amount of equity capital to finance a more extensive portfolio of properties.

Highly-levered REITs face significant challenges that are particularly important during financial crises. Financial leverage puts a firm in a position where it is often forced to go to the capital markets; when doing so, it has the least advantage. Indeed, during the Financial Crisis of the late 2000s, highly levered REITs were forced to roll over their debt in a situation with falling rents, a dysfunctional debt market, and a real estate equity market that had fallen by 70 to 80 percent. They faced two unattractive choices: (1) issue equity at unattractive terms or (2) sell properties into a panicking market. Many of the highly levered REITs did both, resulting in lower returns during the crisis period.

A parallel crisis among second-tier banks in the United Kingdom exacerbated the REIT crisis in the United States. British banks had lent extensively to real estate borrowers based on rising home prices in the late 1960s and early 1970s, underwriting loans that were excessively large relative to the value of the properties that served as collateral for the loans. At the beginning of the 1970s, a collapse in housing prices, coupled with an increase in interest rates, left smaller banks with loans for which the outstanding debt was greater than the value of the homes.

The Bank of England led discussions with lenders, resulting in bailouts of approximately 30 smaller banks and assistance to 30 others. While all banks could pay depositors, the Bank of England lost 100 million pounds.