Skip to main content

Theory and Empirics of Asymmetry

Periodic Reporting for period 2 - TEA (Theory and Empirics of Asymmetry)

Reporting period: 2018-09-01 to 2020-02-29

Why are the financial markets reacting to incoming macroeconomic news the way they do? Do these reactions differ if news are positive or negative, or if the same news arrive in good times or bad? Do markets and the real economy react differently to monetary policy at different times, or when policy is expansionary as opposed to contractionary? These are some of the key questions of the Theory and Empirics of Asymmetry (TEA) ERC grant.
Understanding the behavior of financial markets is important for several reasons. First of all, we would like to think that asset prices embed information about beliefs and expectatons of market participants. Reverse engineering those beliefs and expectations from prices require having a deep understanding of the relationship between macroeconomic fundamentals and asset prices. However, relating asset prices to macro news, even in short windows around macroeconomic data releases was not satisfactorily done until now. It is important to make more headway in this.
Also, regulating financial markets properly requires understanding asset price formation and volatility. Again, having an understanding of what drives asset price changes and whether these are asymmetrically affecting asset prices is important.
TEA focuses on the questions above. We would like to have a better understanding of drivers of asset prices, whether these relationships are state dependent, and whether monetary policy—transmitted through financial markets—itself behaves differently in good and bad times, and whether its effects play out differently.
There are three completed papers and ongoing work. One of the papers introduces new methodology to study the relationship between asset prices and macroeconomic news. The key insight is that while market participants perceive various kinds of news, researchers only see as “news” headline items for which there are surveys. Hence, markets react to more news than is observable to researchers. My research group and I developed a method that estimates what the unobserved news look like, and what the market reaction to these is. When studied using this method, it turns out that macroeconomic data releases explain essentially all of yield curve changes in the intraday windows in which the releases take place. It also turns out that yield curve reactions to data releases are not asymmetric.
Other work, in conjunction with economists at the ECB, an eventstudy database of euro area monetary policy is constructed and market perceptions of ECB policy communication is studied. The database is public and the ECB will keep it updated, this will help monetary policy research in Europe. We learned that market reactions to ECB communication has been remarkably stable but that policy communication itself has changed over time. And, again, we did not find signs of asymmetric reactions of interest rates to policy releases.
The last completed paper focuses on the heterogeneity in stock price reactions to monetary policy. When stock indices go up or down in response policy surprises, not all firms’ stock prices go up or down equally, not even for firms within the same sectors. But why? We relate the difference to firms’ balance sheets and the sensitivity of their cash flows to monetary policy. We devise tests to understand whether stock market participants, on average, know the balance sheets of these firms or use rules of thumb and find that there is remarkable sophistication in stock markets. We then show that firms that have more sensitive cash flows to monetary policy also see their investment more affected by monetary policy. This is very strong evidence for cash flow sensitivity of investment, and helps isolate a transmission channel for monetary policy.
Ongoing work focuses on why market reactions to monetary policy may be time varying, on measuring policy surprises in the US, and on applying the new event study methodology to European data.
The papers described above all represent work that are beyond the state of the art before they were carried out. As described there, this is the case both methodologically, and in terms of the economic insights gained.
The key line of research now focuses on possible asymmetries in financial market reactions to monetary policy surprises, how these relate to various ways in which central banks may be communicating their private beliefs or information to financial markets. I expect to make headway in this regard until the end of the project and be able to better understand the time varying interplay between monetary policy, financial markets, and the real economy.