Aggregate Duration Risk

For the past five years, central banks around the world have pushed down interest rates on traditionally safe assets such as Treasury bonds close to the zero boundary in the hope of sparking economic growth. Eventually, interest rates will rise again as heralded in a recent speech by Chairman Bernanke. Sudden changes in interest rates can have sweeping consequences for asset prices. The goal of the proposed research is to study this relationship and to understand how asset prices react to varying yields. In particular, I plan to measure and analyse the quantity of interest rate or duration risk in the economy both theoretically and empirically.

While a rise in interest rates has obvious implications for investors in bond markets, it also affects other assets and the overall economy in general. For example, a recent report by the International Monetary Fund cautions that the Fed's monetary policy could cause "excessive interest-rate volatility that could have adverse global implications''. While there seems to be widespread agreement that higher interest rates could have negative consequences for a fragile economy, we have surprisingly little knowledge about (i) how to measure aggregate interest rate risk and (ii) to what extent and through which channels it affects the aggregate economy. For example, the supply and the duration of Treasury securities and corporate debt, the duration of pre-payable mortgages, the duration of commodities, and the correlation between yields and the stock market returns all have significant exposure to interest rate risk and at least part of the time-variation in this quantities can be attribute to fluctuations in interest rate risk.

The project will first deliver an aggregate proxy of duration risk in the overall economy by studying the effect of interest rates on a whole cross-section of assets. This is a contribution by itself as such a measure currently does not exist. Being able to quantify the amount of duration is tremendously useful not only for investors or institutions aiming at hedging this risk but also for central bank authorities who have to monitor and supervise this risk and take it into account when conducting monetary policy. In other words, this proxy can answer questions of the form: If interest rates increase by 1%, how and how much do prices in stock, bond or commodity markets react? Or: How does a 1% change in interest rates affect the risk premia or the compensation for risk in these markets?

The data will be available online and kept up-to-date so researchers have access to the relevant data in a timely manner. The research output will be presented at conferences and seminars and submitted for publication in top journals to establish our measures in the academic world.

Part of the research will be undertaken in collaboration with Dr Mueller from LSE, Dr Malkhozov from McGill University and Dr Venter from Copenhagen Business School. We have collaborated in the past and this has already proved very beneficial. Being able to work with researchers outside the UK provides a unique opportunity to maximize the level of impact of the proposed research.

In summary, research in the field of quantifying interest rate risk is very active at the moment and there is a great interest in this work and I personally know and interact with most of the relevant researchers in the field. Towards the end of the award period I plan to organize in collaboration with the LSE Financial Markets Group a specialized workshop and invite the relevant researchers in the field. This will further help disseminate the research to a broader academic audience.

While answering the before-mentioned research questions is interesting from an academic standpoint, it is also relevant from a more practical point of view. Potential beneficiaries outside the academic circle are financial institutions and insurance companies, pension funds, and central banks.

1) Financial institutions and insurance companies have to supervise their interest rate risk as they are regulated by the so-called Basel accord. For example, the new regulation, Basel III, forces banks to hold more capital to guard against changes in interest rates. A detailed knowledge of how large the interest rate risk really is, is therefore of utmost importance.

2) It is well known that pension funds and life insurance companies have a large duration gap, which is mainly driven by the fact that assets have longer duration than liabilities. Understanding how to quantify this risk is vital for their risk management purposes.

3) Central banks can profit in two ways: First, central banks often have a monitoring role for financial institutions. For example, in the UK, the FSA has to monitor the use of interest rate instruments used to hedge interest rate risk. Second, quantifying duration risk allows central banks to better set their monetary policy, i.e., how large should interest rate changes be?

In short, the current environment highlights the need to understand the interplay between hedging of duration risk and first and second moments of different securities' returns. However, it is of general interest to understand how duration hedging activity can affect asset prices, as this will help to build better risk management models and help assessing central bank monetary policy.

An important path to ensuring the impact of the research is to make the aggregate duration risk factor estimates available. Currently, there is no such data publicly available except for the mortgage-backed securities (MBS) market. The proposed research will establish aggregate duration risk to be of equal importance as for the MBS market. Providing convenient access to my data is crucial to a wide dissemination of my results and to ensure that my contribution will get credited in the work of others.

Besides the creation of new data, the main output will be in the form of articles. Academic papers will be submitted (drafts in the earlier stages, completed versions later on) to academic conferences that will be held during and after the completion of the grant period. In particular, I plan to submit to the NBER, CEPR, AFA, EFA and ES meetings, that all have their submission deadlines in February and March 2015. Furthermore, I plan to present my work at more specialized conferences focusing on central bank policy. Given the strong interest in my previous research in MBS duration risk from the financial industry and central banks, I will also submit our work to practitioner conferences such as the Annual Inquire Europe Symposium.

I also seek to present my work in seminars at various academic or central banking institutions (namely the Bank of England, the ECB, the Swiss National Bank, the Federal Reserve Board and the New York and San Francisco Fed) as my measures can help in assessing interest rate risk in the financial markets and hence can help central banks perform their monitoring role. All collaborators on the various projects are well connected in the academic and central banking community and in the past I have presented our work all over the world.

In summary, the proposed research under the award seeks to make both empirical and theoretical contributions to the debate about how interest rates affect prices across different markets. The research is highly topical and relevant for academia, monetary policy, and the financial industry.