Transmission Channels of Financial Shocks to Stock, Bond, and Asset-Backed Markets

An Empirical Model

Massimo Guidolin, Viola Fabbrini, Manuela Pedio

• Publisher: Palgrave Macmillan
• Year: 2015
• ISBN:
  9781137561398
• Topic: Business
• Subtopic: Finanza d'impresa
• Index: Business

Researchers, policymakers and commentators have long debated the patterns through which adverse shocks in a few markets may quickly spread to a range of apparently disconnected financial markets causing widespread losses and turmoil. This book uses modern linear and non-linear econometric methods to characterize how shocks to the yield of risky fixed income securities, such as sub-prime asset-backed or low-credit rating sovereign bonds, are transmitted to the yields in other markets. These include equity and corporate bond markets as well as relatively risk-free fixed income securities, such as highly rated asset-backed securities and sovereign bonds from core Eurozone countries. The authors analyse and compare the results from linear and non-linear models to identify and assess four distinct contagion channels characterizing both US and European financial markets. These include the correlated information, risk premium, flight-to-liquidity, and flight-to quality channels. The results of this study support the theory that both investors and policy-makers ought to pay special attention to liquidity and commonalities in the perceptions of the probabilities of default, as channels through which financial shocks propagate.

1

The Background: Channels of Contagion in the US Financial Crisis

Abstract: In this introductory chapter, we provide the background material that will help a reader to understand the findings presented in the rest of the book. First, we provide a review of the key facts that characterized the onset and the unfolding of the 2007–09 US financial crisis. In addition, we explore different definitions of financial contagion and discuss the key papers published on the topic because they provide benchmarks to our empirical analysis.

Keywords: contagion; cross-asset contagion; financial crisis; sub-prime mortgages

Fabbrini, Viola, Massimo Guidolin, and Manuela Pedio. Transmission Channels of Financial Shocks to Stock, Bond, and Asset-Backed Markets: An Empirical Model. Basingstoke: Palgrave Macmillan, 2016. DOI: 10.1057/9781137561398.0005.

The core of this book is devoted to the discussion of the contagion channels that contributed to the propagation of a shock in the speculative grade asset-backed securities (henceforth, ABS) market to the other asset markets during the recent US financial crisis. Consequently, in this chapter, we provide the necessary background to follow the rest of the analysis. In particular, in the first paragraph, we propose a short synopsis of the main events that have marked the 2007–09 crisis. Our objective is not to exhaustively list all the significant developments or to discuss causes of and solutions to the financial crisis. A number of excellent analyses of these issues have been circulating or are already published; see, for example, Wheelock (2010). On the contrary, we devote special attention to emphasizing the steps through which shocks may have propagated from one market to others during this specific historical event. Indeed, in the second section, we provide a review of the main definitions of financial contagion and a distinction of the different contagion channels that have been explored in the literature. Not only is distinguishing among such channels intellectually rewarding, but also their relative importance and empirical incidence during a specific period of crisis may be highly informative to policy-makers (who manage and fight the effects of the shock) and investors alike.

1.1A brief review of the sequence of events during the US financial crisis

The financial crisis began with a sharp downturn in US residential real estate markets as a growing number of banks and hedge funds reported substantial losses on subprime mortgages and mortgage-backed securities (MBS), the biggest and best-known segment of the ABS market. The crisis had been slowly building up since the early months of 2007. For instance, in late February 2007, the Federal Home Loan Mortgage Corporation (commonly known as Freddie Mac) had announced that it would no longer buy the most risky subprime mortgages and MBS. This meant that a large portion of the process of origination and securitization of subprime MBS would have to be moved over to the private-sector segments of the US residential mortgage market. In April 2007, New Century Financial Corporation, a leading subprime mortgage lender, had filed for Chapter 11 bankruptcy protection. In June 2007, Standard and Poor’s and Moody’s Investor Services had downgraded over 100 bonds backed by second-lien subprime mortgages. However, the first major step towards a spiraling crisis was marked by Fitch Ratings’ decision in August 2007 to downgrade one of the major firms specializing in mortgage intermediation in the subprime segment, Countrywide Financial Corporation. As a result, Countrywide was forced to borrow the entire $11.5 billion available in its credit lines with other banks, which was first-hand evidence that the crisis was destined to spread from the mortgage market to the financial intermediaries backing its operators.

In terms of pricing and trading volumes, the crisis first appeared to be spreading beyond the boundaries of the US mortgage market when it spilled over to the interbank lending market in early August 2007. The London Interbank Offered Rate (LIBOR) and other funding rates spiked after the French bank BNP Paribas announced that it was halting redemptions for three of its investment funds. By wide consensus among researchers and policy commentators (see, for example, Wheelock, 2010), these two negative developments mark an arbitrary and yet useful onset date for the crisis.

Initially, the Federal Reserve’s (henceforth, Fed) reaction was limited to calming markets by reminding banks of the availability of the discount window. This was done by extending the maximum term of discount window loans to 30 days and lowering the Fed fund rate target, initially (between August and September 2007) by 50 basis points.¹ Financial strains eased in September and October 2007, but reappeared in November, because many banks found themselves unable to fulfill their dollar funding needs. In December 2007, the Fed announced the establishment of reciprocal swap currency agreements with the European Central Bank (ECB) and the Swiss National Bank to provide a source of dollar funding to European financial markets, and announced the creation of the Term Auction Facility (TAF) to lend funds directly to banks for a fixed term.²

Despite the relatively small size of the ABS market in the US, the shock rapidly triggered negative and widespread consequences in all credit and, eventually, bond markets (see, for example, Gorton, 2010). The first reason triggering such a chain reaction was the immediate negative response of lenders in the repo market, an episode that has become known as a repo run in the literature (see Gorton and Metrick, 2012).³ Before the crisis, the repo market represented a fundamental source of funding to financial institutions, and a large part of the collateral requirements were met through securitized ABS products. After the shock in the ABS market, lenders became uncertain about their ability to quickly liquidate assets other than Treasuries in the event of default of the counterparty of the repo transaction (see, for example, Adrian, Begalle, Copeland, and Martin, 2013). Therefore, lenders restricted their financing to short-term transactions against Treasury bonds, while severe haircuts were applied on other assets, because of their modest liquidity (see, for example, Hördahl and King, 2008).⁴ Following the changes in the conditions set by lenders on repo transactions, investors’ demand for Treasury bonds rapidly increased, bringing the yield on this asset class as well as the Treasury repo rate to levels close to zero.⁵ After that and throughout the crisis, the repo activity was, in fact, mainly driven by the need to borrow Treasuries that were scarce in the market, rather than to obtain financing. Moreover, the liquidity shock in the repo market forced financial institutions to fire sale their asset holdings in order to raise money, so that the shock to the ABS market rapidly spread to the corporate bond and stock markets.⁶ To address this type of market freeze, in March 2008, the Federal Reserve established the Term Securities Lending Facility (TSLF) to provide secured loans of Treasury securities to primary dealers for 28-day terms, and the Primary Dealer Credit Facility (PDCF) to provide secured overnight loans to primary dealers under Section 13(3) of the Federal Reserve Act, which permits the Federal Reserve to lend to any individual, partnership, or corporation “in unusual and exigent circumstances”.

The crisis intensified during the final months of 2008. Lehman Brothers, a major investment bank, filed for bankruptcy on September 15. Lehman’s bankruptcy produced an immediate fallout. On September 16, the Reserve Primary Money Fund announced that the net asset value of its shares had fallen below $1 because of losses incurred on the fund’s holdings of Lehman commercial paper and medium-term notes. The announcement triggered widespread withdrawals from other money funds, which prompted the US Treasury Department to announce a temporary program to guarantee investments in participating money market mutual funds, the Asset-Backed Commercial Paper Money Market Mutual Fund Liquidity Facility (AMLF), set up to extend non-recourse loans to US depository institutions and bank holding companies.⁷

In spite of the strength of these massive policy interventions, financial markets appeared to be mired in a state of persistent turmoil, worsened by the almost complete freezing of the commercial paper and ABS markets, where the activity of issuance and origination had completely stopped in the aftermath of the severe losses imposed by Lehman’s default. In particular, while the policy interventions led to some beneficial effects on the short-end of the fixed income markets, the situation remained difficult in most other segments, especially as far as ABS and associated derivative products (for example, collateralized debt obligations written on portfolios of MBS) were concerned. The result was that immense portfolios at several multinational financial institutions remained extremely illiquid and potentially exposed to tremendous losses upon what are often called fire sales. In November 2008, the Fed announced the creation of the Term Asset-Backed Securities Lending Facility (TALF). Under this facility, the Federal Reserve Bank of New York provided loans on a non-recourse basis to holders of AAA-rated asset backed securities and recently originated consumer and small business loans. At the same time, the Federal Open Markets Committee (FOMC) announced its intention to purchase large amounts of US Treasury securities and mortgage-backed securities issued by Fannie Mae, Freddie Mac, and Ginnie Mae (the FOMC was to increase the amount of its purchases in 2009). These interventions helped to propagate the strains to the long-term Treasury market, where prices were lowered by the combined thrust of negative expectations on the economic outlook and the effects of the Fed’s purchases.

In early 2009, fears spread that the enormous market for securitized commercial mortgages was on the brink of collapse, similarly to the subprime residential mortgage market in the spring of 2007. Stock markets were severely affected, with heavy losses that – purely for accounting reasons, as firms are slow to adjust dividend pay-outs, which were still to drop, as the US economy had just timidly entered a recession – implied that dividend yields had shot up. The policy-makers explicitly admitted that financial markets remained strained, and they decided that the extraordinary measures enacted between December 2007 and December 2008 should be extended for as long as necessary.

The turnaround and the exit from the crisis seem to have occurred – we claim now in hindsight – between the late spring and the fall of 2009. In fact, while in June 2009 the Fed had still announced a number of extensions and modifications to a number of its liquidity programs, a novel desire to fine-tune the programs had replaced the drive towards expanding them that had dominated policymaking until April 2009. With the situation rapidly improving, and the short-term debt (especially interbank) markets going through an unfreezing cycle opposite to the severe, paralyzing disruptions experienced in September–November 2008, in November 2009 the Fed approved a first reduction in the maximum maturity of the credit it offered through the discount window. Although the discount window never played a major role in the credit-easing policies of the Fed, this represented the first official acknowledgement that the financial system was healing and the crisis possibly ending. This was made clear not only by the Fed but by all central banks around the world when – between late 2009 and early 2010 – they ended some or most of the public support measures introduced in response to the financial crisis. For instance, the Fed completed its purchase of Treasury securities in October 2009.⁸ In the same month, the ECB conducted a last 12-month euro repo to finance banks, and the Bank of Japan stopped its purchases of commercial paper and corporate bonds; finally, the Swiss National Bank ceased providing Swiss francs through foreign exchange swaps against euros in January 2010. On the demand side, the take-up of many measures drastically declined around the turn of the year. This seems to reflect better market access and hence reduced demand for government support. In fact, Guidolin and Tam (2013) use data on US bond yields and spreads to date the end of the US financial crisis to between June and December 2009.

However, what we have offered represents a simple narrative account of the US financial crisis, and its power to spread from the ABS market to all other financial markets was simply presumed. Moreover, specifically how these spillover effects took place remains interesting. The next section analyzes the definition of contagion and reviews the relevant literature, while Chapters 4 through 6 use modern statistical techniques to tackle exactly the questions we have raised above.

1.2Modeling alternative cross-market contagion channels

Before analyzing the different contagion channels, it is useful to review the definition of the phenomenon. The literature provides a number of alternative and yet complementary definitions of financial contagion. Kyle and Xiong (2001) describe contagion as an episode of declining asset prices, tightening of liquidity conditions, and increased volatility and correlations, which rapidly propagates from one market to another. Hence, besides the dynamics of first- and second-order moments, which has also been discussed elsewhere in the literature (see below), the lack of liquidity and its propagation across markets would matter. Dornbusch, Park, and Claessens (2000), Kaminsky, Reinhart, and Végh (2004), and Longstaff (2010) define contagion as an episode of significant increase in cross-market linkages, following a shock to one market. In this perspective, the strength of cross-market connections is the defining feature of contagion. In Pritsker (2001), the most salient trait of contagion is that the negative effects generated by a shock to one market on the value of assets traded in other markets cannot be explained by changes in the fundamentals characterizing these other markets.

These definitions of the general concept of contagion have been subjected to a number of detailed applications that have further revealed the general nature of the phenomenon. For instance, a large number of existing studies have focused on cross-country contagion to identify episodes of international crisis spillovers, which arise when a shock to one national market triggers significant and immediate financial effects in other countries (see, for example, Kaminsky, Reinhart, and Végh, 2004; Allen and Gale, 2004; Kodres and Pritsker, 2002; King and Wadhwani, 1990). However, in light of the recent subprime crisis, recently researchers have also paid increasing attention to the empirical analysis of cases of cross-asset contagion, that is, within-country, cross-market contagion episodes (see, for example, Longstaff, 2010; Guo, Chen, and Huang, 2011). For instance, scores of recent articles have aimed to identify the contagion channels that are typically active during national, closed-economy financial crises (see, for example, Kodres and Pritsker, 2000; Brunnermeier and Pedersen, 2009; Vayanos, 2004; Caballero and Kurlat, 2008; Longstaff, 2010).

Recently, researchers have achieved progress by isolating and trying to measure the strength of four distinct propagation channels: the correlated infor...

Table of contents

• Cover
• Title
• 1  The Background: Channels of Contagion in the US Financial Crisis
• 2  Methodology
• 3  The Data
• 4  Estimates of Single-State VAR Models
• 5  Results from Markov Switching Models
• 6  Estimating and Disentangling the Contagion Channels
• 7  Comparing the US and European Contagion Experiences
• 8  Conclusions
• References
• Index