Subprime Crisis, The: Lessons For Business Students

Lessons for Business Students

Jaime Luque

• Publisher: WSPC
• Year: 2017
• eBook ISBN: 9789813200050
• Topic: Economics
• Subtopic: Finance
• Index: Economics

-->

The first major global economic contraction of the 21st century, or the "Great Recession", as it is more commonly known, began in December of 2007, and would prove to be one of the most catastrophic economic events in postmodern history. The bursting of the American housing bubble and the subprime mortgage crisis that followed are widely attributed as being the primary causes for this economic downturn.

The content of this book presents an in-depth analysis of several of the leading causes of the subprime crisis, as well as the subsequent measures that were used to contain a widespread economic recession. It is the aim of this book to provide adequate information and tools for readers to gain insight on how we can prevent the same mistakes from happening again. This analysis also explores an interesting question when considering public policy; was the economic disaster, that ensued, caused by overzealous consumers who borrowed more than they could afford, or was it caused by institutional entities that gambled away the American economy?

-->

Part I

Overview of the Subprime Mortgage Crisis

Chapter 1

Understanding the Subprime Crisis

In collaboration with Thomas Sullivan and Jeremy Scheer

It is often said that, “hindsight is 20/20”, a saying which rings especially true when considering an event such as the Subprime Crisis. How could it be that economists, Wall Street executives, financial analysts, and even Ben Bernanke, the chairman of the Federal Reserve at the time, didn’t see an event of this magnitude coming? In this chapter, we investigate economic indicators leading up to the subprime crisis, and we try to uncover if warning signs existed that would point to the imminent fall of the housing market. Answering these questions is crucial in shaping our understanding of how markets work, which in turn will enable us to identify warning signs to prevent events like this from happening again.

A repeated common argument during the Great Recession was that, if provided with perfect foresight on home price depreciation, economists could have predicted a substantial increase in the foreclosure rates that actually occurred. However, economists did not possess this foresight and believed the likelihood of a countrywide meltdown was highly unlikely, and therefore were unable to predict the consequences that home price depreciation had on foreclosures.

The notion that one factor could have such a profound destructive impact on an entire economy may appear astounding on the surface. When considering the U.S. economy, one that is led by exceptionally knowledgeable individuals with superior technology and research techniques, this viewpoint becomes even more unclear. What were the ex-ante projections of economists prior to the collapse? Could one deteriorating factor destroy an entire economy?

This chapter focuses on the destructive effect that the decline in home prices initiated as well as factors that must be taken into account for accurate analysis. The periods under consideration were before the crisis when home prices were skyrocketing, the peak of the market in mid-2006, and the decline thereafter. Our discussion is based on the seminal work by Gerardi et al. (2010), who attempted to determine to what magnitude investors could have foreseen the escalation of foreclosure rates and the degree that various factors played in this determination. This is an arduous task due to the fact that there are so many elements to take into consideration when examining the period leading up to the subprime crisis, as well as the crisis itself. They attempted to assert their opinion that borrowers were the most sensitive to home price depreciation, and that this price decline was the factor that actually led to the collapse of the economy. They recognized factors such as underwriting standards, leverage, documentation in lending, risk layering, and a variety of other aspects in formulating their argument, although they depicted these factors as marginal in comparison to the decline of house prices. Gerardi et al. collectively achieved their goal by using two samples; Massachusetts data starting in the late 1980’s to capture a time of extreme house price declines prior to the Great Recession and a nationwide sample that inspects the state of the economy in the years leading up to the Great Recession. By using contrasting samples and time periods of when these models transpired, they were able to clearly identify what they believed to be the leading cause of the proliferation of defaults (decline in house prices). With this knowledge, they could assess if it was possible for economists to prevent this catastrophe beforehand.

The information regarding mortgages sold into private label market backed securities comes from the TrueStandings Securities ABS data, which is provided by First American Loan Performance. This national data is commonly used in the industry to follow the performance of mortgages in mortgage-backed securities (MBS), and it would have been used before and during the subprime crisis. Furthermore, Gerardi et al. (2010) restricted this data set specifically to the three most popular types of loans at the time: fixed interest rates to maturity and two types of loans at adjust called the 2/28s and 3/27s. To add to this data set, they also used data from the Standard & Poor’s Case Shiller Home Price Index, state-level house price data from the Office of Federal Housing Enterprise Oversight (OFHEO), state-level unemployment rates, monthly oil prices, and interest rates. Additionally, they used data from the Census Bureau on ZIP code level data on average household income, share of minority households, share of households with a high school education or less, and the child share of the population. All of this data is used in order to have a more complete picture of the situation before they modeled their results.

Gerardi et al. (2010) also used a significant amount of data from the state of Massachusetts. This is publicly available information, and contains individual-level data on both housing and mortgage transactions in the state of Massachusetts, from county-level registry of deeds offices. Additionally, they used information from the Warren Group, which has followed home buyers in Massachusetts since the late 1980s. This is ownership-level data, which can include more than one mortgage loan since it spans the homeowner’s time at the specific property. Even though this data was not widely used by the industry, it was available information before and during the subprime crisis and contains relevant information.

Expectations for the future of home prices are one of the driving forces in the subprime crisis. If home prices continued to appreciate, things would have been fine. However, the rate at which home prices were appreciating was unsustainable in the long-term, and eventually when the downturn came, a lot of homeowners were not in a situation where they could ride out the decline without having to foreclose on their homes. The model developed in Gerardi et al. (2010) aggregated all of the data gathered and tried to predict what analysts would have been able to predict prior to the Great Recession. These models proved to be pretty accurate and showed that analysts could have seen this decline in home prices coming; they just thought it was improbable.

In order to see what the financial community was thinking at the time, a combination of different parts of reports written by analysts at the five major banks, J. P. Morgan, Citigroup, Morgan Stanley, UBS, and Lehman Brothers was used. From these reports, there were some underlying themes that provided insight on the crisis. One of the main points discovered was that many analysts anticipated the possibility of a crisis in a qualitative way, even discussing potentially different ways in which it could happen, but the analysts never made measurable implications. Another important take away that adds to their main point is that analysts were remarkably optimistic about home price appreciation, which directly affected their actions. In 2006, investors expected that household prices would continue to appreciate or to plateau, not to start dropping, even though some data showed the possibility of a crash. Analysts understood that a major fall in home price appreciation would lead to a dramatic increase in problems in the subprime market, but they simply thought that a 20 percent nationwide fall in prices was impossible. These analysts chose to focus on scenarios that gave the lowest default rates, which directly affected their outlook for subprime mortgage performance.

Gerardi et al. (2010) main point is to prove that reduced underwriting standards alone cannot explain the dramatic rise in foreclosures. Even though it was a part of it, it was not the only cause. Although newly originated subprime loans carrying less than full documentation did increase from 1999 to 2006, they were by no means the majority of the business, nor did they increase dramatically during the credit boom.

With all of their data from ABS, S&P/Case-Shiller, and Office of Housing Enterprise Oversight, Gerardi et al. could estimate what an analyst with perfect foresight about home prices, interest rates, oil prices, and other variables would have predicted for prepayment and foreclosures in 2005–2007. By using the data, they created a model to predict the number of foreclosures in the fourth quarter of 2007 and then compared them to the actual numbers. The model that used data from 2000 to 2004 does very well at predicting foreclosures, accounting for approximately 85 percent of cumulative foreclosures in the fourth quarter of 2007. The model that used 2005 data did not perform as well, but it still predicted 63 percent of cumulative foreclosures in the fourth quarter of 2007. There were significant differences in the performance of the model when using the data from the two sample periods, but both can predict a majority of the foreclosures that occurred. This shows that if investors been endowed with perfect foresight about actual home price changes, then they could have predicted a significant portion of the increase in foreclosure rates that occurred, although not all of it.

The decline in home prices and housing equity were the key drivers of the foreclosures. Other factors such as underwriting standards did not deteriorate enough to explain the foreclosure. To further test how underwriting affected the housing crisis, they examined the different types of amortization schedules used at the time. Non-traditional amortization schedules started to become increasingly popular among subprime loans, and highly leveraged loans increased as well, growing from pretty much 0 percent in 2001 to close to 20 percent of subprime originations by the end of 2006.

It can be deduced that the presence of home price depreciation and diminished housing equity were the foundation of the rise in foreclosures. By no means did Gerardi et al. disregard related factors, although they attribute the collapse mainly to the fall in home prices nationwide. In 2000, high combined loan to value (CLTV) ratio lending accounted for around 10 percent of loan originations, which rose to over 50 percent by 2006. The notion that a growing amount of subprime loans were provided to those with high loan to value (LTV) ratios does increase the probability of default, although not to the point where it should be considered a more vital factor than home price depreciation. From 2004 to 2008, loans without a second lien had an average CLTV ratio of 79.9 percent, compared to those with a second lien had an average CLTV ratio of 98.8 percent. This indicates the use of risk layering, which similarly to the amount of high LTV ratio loans given in this time, increased. Risk layering is the process of creating loans with a combination of risk factors, such as high leverage and minimal documentation. This risk layering created a situation where the lender did not have accurate knowledge on the borrower’s background. Therefore, loans were provided to individuals that would otherwise be considered too risky. They explained that documentation must be taken into account due to the fact that risk layering is commonly a result of low loan documentation, and a high CLTV ratio. During this time, documentation of loans was shaky at best, therefore providing the borrower with an opportunity to exploit the lender’s services. The combination of high leverage, a high LTV ratio, poor documentation and risk layering led to a false degree of certainty that these borrowers would not default on their loans. Gerardi et al. took factors other than home price depreciation into account when assessing the influences of such a rapid increase in the rate of foreclosures, although they regard these factors as secondary with respect to the effect they had on the crisis as a whole.

Gerardi et al. (2010) ultimately concluded that economists could have predicted the reactiveness of foreclosures to home price depreciation. Their analysis indicated that they were not able to predict this reactiveness due to the optimistic view that the future value of home prices would not dramatically drop. In one example, one of the five major banking institutions mentioned earlier provided an accurate representation of what would happen in the most disastrous of situations. It stated that, “forecasting delinquencies in May 2008 with a 20 percent fall in house prices (roughly what happened), would have predicted a 35 percent delinquency rate and a 4 percent cumulative loss rate. The actual numbers for the 2006-1 asset-backed securities index (ABX) were a 39 percent delinquency rate and a 4.27 percent cumulative loss rate.” These numbers are comparable to what actually happened when the recession hit, ultimately proving that economists knew the potential effects of such home price depreciation, but did not take into account the likelihood of a meltdown actually occurring. Therefore, it can be concluded that analysts did not believe such extreme home price depreciation could occur, for it they did, the foreclosures that followed could have been potentially avoided.

Mian and Sufi (2014) both support and further illustrate Gerardi et al. (2010) viewpoints on the housing crisis. For example, Mian and Sufi acknowledged that before the recession purchasing on credit became more acceptable, consumer’s debt as percentage of household income sharply increased before the crisis, and that there was a large drop in household spending once the crises started. Although all of these factors do not directly support Gerardi et al. conclusions, they parallel the factors that occurred before the recession, and the effects it had during it.

Another connection between Mian and Sufi (2014) and Gerardi et al. (2010) is that both studies touch on the increased gap between the rich and the poor that became apparent as a result of high debt and the dramatic decline in house prices. Similarly, Mian and Sufi (2014) recognized the decline in housing prices as the driving force, although they place more emphasis on the dichotomy between the rich and the poor than do Gerardi et al. (2010).

One potential problem in Gerardi et al. (2010) is that they had to close their data set on December 2007, so they only were able to track mortgages originated through December 2006. It would be helpful to see this data throughout the entire financial crisis. Obser...

Table of contents

• Cover Page
• Title
• Copyright
• Dedication
• Preface
• About the Editor
• Contents
• Part I: Overview of the Subprime Mortgage Crisis
• Part II: Foreclosures
• Part III: Government Intervention
• Part IV: House Prices and Lending Standards
• Bibliography
• Index