Behavioral Finance and Financial Stability

About This Research

The Behavioral Finance and Financial Stability Project, founded at the Harvard Business School, supports research collaborations between faculty and students across Harvard University to understand, predict, and prevent financial instability. The BFFS project also maintains an ongoing real-time database of financial stability and investor sentiment measures that it makes available to researchers and the broader public. Please visit our About page for more information about the project.

→ Proceed to Data Section

Research Highlights

Figure
Crashes and non-crashes after a 100% runup in the us industry sectors.

Bubbles for Fama (BFFS WP #011)

Robin Greenwood, Andrei Shleifer, and Yang You
FEB 2017

Nobel Laureate Eugene Fama once famously stated that stock markets do not exhibit price bubbles. Evidence supporting this claim boils down to the argument that past run up in stock prices do not seem to predict lower future returns. Authors Greenwood, Shleifer, and You seek to evaluate Fama's claim using stock return data gathered from a variety of US industries, and a gamut of international stock market sectors. They find that while Fama is correct in asserting that sharp price increases do not predict lower returns going forward, these increases do predict substantial heightened probability of a crash. Simple attributes related to the price run up can help predict both the crash probability and future returns.

See Robin’s other research here and Andrei’s other research here.

Related Themes: Measuring Sentiment & Expectations

The Fed, the Bond Market, and Gradualism in Monetary Policy (BFFS WP #010)

Jeremy C. Stein, and Adi Sunderam
Feb 2017

Fed watching has become a routine business for bond market participants and the financial press. Monetary policy announcements are analyzed word by word for clues to the central bank's future actions. Such scrutiny seems inconsistent to interest rate setting models in which policy makers target specific macro variables. If the Fed is simply responsive to publicly observable variables, then the bond market would only react to the release of these variables and not to the policy announcements. Stein and Sunderam posit that private information and preference for volatility smoothing of long term bond yields by the Fed underly this interaction between the Fed and bond market participants. Policy makers are gradual in setting its short term yields to smooth long term yield volatility. However, market participants anticipate this gradualism. In equilibrium, long term yields stay volatile. The authors derive several normative suggestions for policy makers.

Related Themes: Credit Markets,  Monetary Policy and Money Markets

The Financial Regulatory Reform Agenda in 2017 (BFFS WP #009)

Robin Greenwood, Samuel G. Hanson, Jeremy C. Stein, and Adi Sunderam
FEB 2017

In this note, researchers from the Behavioral Finance and Financial Stability Initiative evaluate the regulatory reforms since the financial crisis. The authors highlight several areas in which the reform agenda has made clear progress. These include heightened capital and liquidity requirements, more stress testing and capital planning at banks, and increased regulatory authority and tools for financial dissolution. Likewise, the authors also note several areas in which reforms should be rolled back or modified.

Related Themes: Monetary Policy and Money Markets,  Stabilization Policy & Regulation,  Size and Growth of the Financial Sector

Figure
Graph shows the size and the composition of assets on the Fed Balance sheet from 2007-2015.

How Quantitative Easing Works: Evidence on the Refinancing Channel (BFFS WP #008)

Marco Di Maggio, Amir Kermani, and Christopher Palmer
DEC 2016

This BFFS working paper examines the empirical effects of unconventional monetary policy. Using novel data on mortgage refinancing, the authors argue that the rounds of quantitative easings had variable results depending on the type of asset purchased and the degree of segmentation for that asset market. QE1, which targeted mortgage markets, had significant effects on aggregate demand and consumption. QE2, which targeted treasuries, had more muted effects.

See Marco's other research here, Amir's other research here, and Christopher's other research here

Related Themes: Credit Markets,  Monetary Policy and Money Markets,  Stabilization Policy & Regulation

Figure
This figure presents the fraction of operational risk weighted asset (RWA) as a percentage of total RWA for 30 GSIBs as disclosed in regulatory filings and investor reports.

Rethinking Operational Risk Capital Requirements (BFFS WP #006)

Peter Sands, Gordon Liao, and Yueran Ma
DEC 2016

In a BFFS working paper, Sands, Ma, and Liao assess the latest Basel Committee proposals for reform on operational risk capital. The authors conclude that neither the existing Basel II framework nor the reforms that are designed to replace it are effective in minimizing negative externalities from operational risk events. The authors make recommendations for an alternative approach.

Related Themes: Stabilization Policy & Regulation

Credit Migration and Covered Interest Rate Parity (BFFS WP #007)

By Gordon Liao
DEC 2016

In this paper, Gordon Liao studies the joint determination of two economically large and persistent violations of the Law Of One Price (LOOP) in the global corporate credit market and the foreign exchange rate market. Post-crisis regulatory and intermediary frictions have severely impaired arbitrage in the exchange rate and credit markets each on their own, but capital flows, such as international debt issuance, bundle together the two LOOP violations. Limits of arbitrage spill over from one market to another.

See Gordon's other research here

Related Themes: Credit Markets

Credit Cycle Feedback Loop
Credit Cycle Feedback Loop

A Model of Credit Market Sentiment (BFFS WP #002)

Robin Greenwood, Samuel G. Hanson, and Lawrence Jin
AUG 2016

Over the past decade, it has increasingly been recognized that investor beliefs play an important role in driving the credit cycle. In "A Model of Credit Market Sentiment", Robin Greenwood, Sam Hanson and Lawrence Jin (of Caltech) develop a model to explore the feedback between credit market sentiment and credit market outcomes. Their model is able to capture many documented features of credit booms and busts, including the link between credit growth and future returns, and the "calm before the storm" periods in which fundamentals have deteriorated but the credit market has not yet turned.

See Robin’s other research here, Samuel’s other research here, and Lawrence’s other research here

Related Themes: Measuring Sentiment & Expectations,  Credit Markets

Monetary Policy and Global Banking (BFFS WP #004)

Falk Bräuning and Victoria Ivashina
JUN 2016

Global banks primarily receive funding in their domestic currencies, but operate and invest in multiple foreign currencies. These cross country lending/borrowing operations create demand for currency hedges. A tightening of domestic monetary policy may drive a bank to lend more in a foreign currency by its effect on the these currency hedges. Falk and Victoria show evidence supporting this channel of cross border shock transmission from data sets of country level firm-claims and individual level syndicated loans.

See Falk's other research here. See Victoria’s other project-related financial stability research here, or her homepage here.

Related Themes: Credit Markets,  Monetary Policy and Money Markets

Graph
The figure shows the return on assets of highly levered (dashed line) and low leverage (solid line) banks from 2003-2015. Contrary to popular opinion, highly levered banks were less profitable than low leverage banks pre crisis.

The Private Costs of Highly Levered Banks

Juliane Begenau and Erik Stafford
JUN 2016

The choice of relatively high leverage as a means to reduce the overall cost of capital by many banking practitioners is interpreted as folly by some, and defended by others as a clever decision given the capital market’s apparent failure to appreciate the associated risks. Begenau and Stafford suggest and provide evidence for an additional consideration: that high leverage signals an excessive focus on financing decisions over good management, specifically over value-relevant investment and operating decisions.

See Erik’s other research here and Juliane’s other research here.

Related Themes: Global Crisis and Debt Data

Does Reserve Accumulation Crowd Out Investment

Carmen Reinhart, Vincent Reinhart, and Takeshi Tashiro
MAY 2016

In this paper, Reinhart, Reinhart, and Tashiro study nine Asian economies in an effort to understand why much of the decrease in investment that occurred during the 1997-1998 crisis has persisted in the years between 1998 and 2014. Unlike Latin American economies during the mid- to late-1980s, investment in these economies does not appear to be suffering at the expense of private consumption or capital flight. For this reason, the authors contend that the traditional concepts of crowding out and leakages must be redefined to be more encompassing.

See Carmen’s other research here and Vincent’s other research here.

Related Themes: Global Crisis and Debt Data,  Monetary Policy and Money Markets

Connectedness and Contagion

Hal Scott
MAY 2016

Systematic runs on financial institutions were the main culprits of the financial meltdown of 2008, not over-exposure in connected balance sheets. Hal Scott argues this insight in his new book, "Connectedness and Contagion." In fact, contagion, as caused by these systematic runs of short-term creditors, still imposes great risks on the global financial system. Hal warns that recent legislative efforts by the US congress have weakened the ability of regulatory bodies to adequately combat contagion.

See Hal's other research here

Related Themes: Global Crisis and Debt Data,  Stabilization Policy & Regulation

Chart
The figure shows the fraction of cov-lite loans by quarter

Covenant-Light Contracts and Creditor Coordination

Bo Becker and Victoria Ivashina
31 MAR 2016

Leveraged loan markets go through episodes in which the typical new loan is issued with far fewer protections to the lender, known as “cov-lite.” Many have pointed to cov-lite issuance as a proxy for credit market overheating. In the first empirical analysis of this topic, Bo and Victoria evaluate whether this development can be attributed to market overheating, increased borrower demand for cov-lite loans, or a rise in creditor coordination costs.

See Victoria’s other project-related financial stability research here, or her homepage here. Data related to this study is publicly available here.

Related Themes: Measuring Sentiment & Expectations,  Credit Markets

Global Cycles: Capital Flows, Commodities, and Sovereign Defaults, 1815-2015

Carmen Reinhart, Vincent Reinhart, and Christoph Trebesch
FEB 2016

Empirical works exploring the relationship between capital flows and economic crisises have been limited by data to several episodes in the modern era. Reinhart, Reinhart, and Trebesch explore the rich history of booms and busts in capital flow by uncovering data from sources going back to 1815. The pattern uncovered by the authors have strong implications on the vulnerabilities of many emerging economies today.

See Carmen’s other research here, Vincent’s other research here, and Christoph’s other research work here.

Related Themes: Measuring Sentiment & Expectations,  Global Crisis and Debt Data

Forward Guidance and the Yield Curve: Short Rates versus Bond Supply

Robin Greenwood, Samuel G. Hanson, and Dimitri Vayanos
07 DEC 2015

The term ‘forward guidance’ – when the central bank guides market expectations – is normally used in reference to central bank policy on short rates. However, quantitative easing (or QE) – the other primary monetary policy tool being used since 2008 – also involves some degree of forward guidance as well. In this paper, Greenwood, Hanson, and Vayanos build a no-arbitrage model of the yield curve that allows for a characterization and comparison of the effects of forward guidance on short rates and forward guidance on quantitative easing.

See Robin’s other research here, Samuel’s other research here, and Dimitri’s other research here.

Related Themes: Monetary Policy and Money Markets,  Stabilization Policy & Regulation

Extrapolation and Bubbles

Nicholas Barberis, Robin Greenwood, Lawrence Jin, and Andrei Shleifer
SEP 2015

At the heart of standard narratives of historical asset bubbles is a high degree of extrapolation – the formation of expected returns based on past returns – by investors. Yet, nearly all bubbles are also defined by very high trading volume, a feature that excessive extrapolation cannot explain alone. Barberis, Greenwood, Jin, and Shleifer develop a novel model of bubbles, wherein extrapolative investors weigh two opposing signals, in an effort to reconcile these two defining features of bubbles.

See Nicholas’s other research here, Robin’s other research here, Lawrence’s other research here, and Andrei’s other research here.

Related Themes: Measuring Sentiment & Expectations

Dollar Funding and the Lending Behavior of Global Banks

Ivashina, Victoria, David S. Scharfstein, and Jeremy C. Stein
AUG 2015

Foreign banks play large roles in the US domestic funding market. However, unlike domestic banks, which are funded by insured deposits, these institutions are funded either by uninsured domestic commercial papers or by insured foreign denominated deposits which are then swapped in the FX market. A shock that forces these banks to switch from commercial papers to deposits can have large consequences on the covered interest rate parity relationship given limited arbitrage capital.

See Victoria’s other project-related financial stability research here, or her homepage here. See Jeremy’s other research here.

Related Themes: Credit Markets

Graph
The figure shows the average premium of short-term T-bills, by week-to-maturity. Short-maturity T-bills have very low yields.

A Comparative Advantage Approach to Government Debt Maturity

Robin Greenwood, Samuel G. Hanson, and Jeremy C. Stein
AUG 2015

Using a novel model that incorporates monetary benefits that investors derive from holding riskless securities, Greenwood, Hanson, and Stein examine how a government should optimally determine the maturity structure of its debt. They explore the results of their model under multiple scenarios depending on whether a government can directly internalize these monetary benefits and on the presence of private sector competition in the production of riskless, money-like claims.

See Robin’s other research here, Samuel’s other research here, and Jeremy’s other research here.

Related Themes: Credit Markets,  Monetary Policy and Money Markets,  Stabilization Policy & Regulation

Figure
The authors recommend opting for treasury rates over the LIBOR as the reference rate for many interest rate derivatives

Reforming LIBOR and Other Financial Market Benchmarks

Darrell Duffie and Jeremy C. Stein
JAN 2015

Duffie and Stein review the history and the role of LIBOR and similar benchmarks used in the modern financial market. The authors argue that the incentive to distort these benchmarks is severe given the benchmarks' polling nature and the sheer volume of the derivative market linked to these rates. The authors make several recommendations on how changes to benchmark definition and the adoption of new overall regulatory policies, can reduce the susceptibility of reference rates to manipulation.

Related Themes: Credit Markets,  Stabilization Policy & Regulation,  Size and Growth of the Financial Sector

→ View More BFFS Research