Does Central Bank transparency impact financial markets? A cross-country econometric analysis.

1. Introduction

Does the degree of information that a central bank releases to the public have any effect on the functioning and efficiency of financial markets? Is there a significant difference between the Federal Reserve announcing policy decisions at the time of Federal Open Market Committee (FOMC) meetings and the Federal Reserve forcing the public to ascertain policy decisions through subsequent movements of the federal funds rate? Some authors, including Blinder et al. (2001), Poole, Rasche, and Thornton (2002), and Chortareas, Stasavage, and Sterne (2002), argue that transparency both helps establish monetary policy credibility in the public's eyes and transfers clearer information to financial markets. Many prominent central banks, including the Bank of Canada, Bank of Japan, and Bank of England, moved towards greater transparency in the 1990s, with the U.S. Federal Reserve System following suit in 1994. However, the European Central Bank (ECB) has resisted implementing openness to the same degree, citing the need for speaking with a single, clear policy voice in contrast to a mix of arguments and opinions from the members of its monetary policy authority. (1)

In fact, is it possible that central banks have perhaps become too open in their discussions with the public, and that a partial return to the days of the Fed temple (2) is warranted? Consider the U.S. experience in 2003. Following a protracted U.S. economic slowdown and a nominal federal funds rate dropping to 1.25%, on November 21, 2002, Federal Reserve officials first suggested using long-term Treasury bond buybacks as a way to help lower long-term market interest rates. However, on July 15, 2003, in his semiannual report to the U.S. Congress, Federal Reserve Chairman Alan Greenspan recanted the hypothesized proposal, announcing that the action was highly unlikely. Bond markets responded emphatically, with the 10-year Treasury bond yield rising 20 basis points that day to 3.94%. Thus, public statements by the central bank moved U.S. financial markets, not because of changing economic conditions, but because of the revision of a publicly declared proposal, which in turn adjusted agents' future expectations. The corresponding volatility in bond markets would have been avoided under a system of reduced openness. Are the occasional public missteps by central bankers, then, worth the increase in public information gained through transparency? Fed officials are so concerned about this issue that in September 2003, the FOMC for the first time held a special meeting to discuss how to communicate effectively with the public.

When the U.S. Federal Reserve System made a move towards greater transparency in the conduct of monetary policy in February 1994, it began announcing its targets for the federal funds rate on the afternoon of FOMC meetings. Previously, the Fed did not announce its policy decisions until six weeks after the meeting. This left the public to guess at the Fed's actions, either by studying economic indicators or by watching the federal funds rate in the days and weeks following FOMC meetings, which led some economists and the media to label U.S. monetary policy as being conducted under "a veil of secrecy." Gaining accurate predictions of Fed policy was so important that an entire industry of "Fed watchers" developed. After 1994, then, was there a noticeable change in the dynamics of U.S. financial markets? Did the degree of uncertainty in interest rate movements drop after 1994 in response to the additional information released by the Federal Reserve System? Starting with the Reserve Bank of New Zealand, other central banks underwent similar institutional reforms throughout the 1990s (Table 1). How did their financial markets react, if at all, to the reduction in informational asymmetries between central banks and the public in these countries? Finally, let us consider countries in which the central banks resisted the trend towards greater openness. Did financial markets react in a similar manner to those in other countries, implying that other forces were at work in changing the nature of financial markets worldwide, or are there inherent differences between financial markets in countries that made the move to greater central bank openness compared to those that kept the same levels of transparency?

Focusing on a set of seven industrialized countries, we study whether selected central banks' move toward more open disclosure during the 1990s improved or worsened the predictability of the corresponding national financial markets. Using both threshold ARCH and vector autoregression frameworks, we find that for all countries except Germany, the forecasting error has decreased for interest rates on the respective government bonds across most maturity lengths, and that the expectations hypothesis has performed better at the short end of the yield curve. For central banks that made the move to greater disclosure, this effect was slightly stronger than those banks that resisted increasing the public's information set. Furthermore, both conditional and unconditional volatility dropped for both groups of central banks. These results are consistent with Tabellini's (1987) view that increased central bank openness removes an extra source of uncertainty, helping the smooth functioning of financial markets.

This study adds to the existing literature in the following ways. First, in contrast to many previous empirical studies involving transparency, such as Chortareas, Stasavage, and Sterne (2002) and Cecchetti and Krause (2002), we examine how the move to greater transparency has influenced financial market factors rather than macroeconomic factors. More specifically, we focus on how increased central bank openness may impact the expectations hypothesis, and thus the term structure of interest rates. Understanding the term structure is of prime importance to central banks, since they can most directly influence short-term rates, yet aggregate demand depends chiefly on long-term interest rates. The effectiveness of the monetary transmission mechanism, then, may be linked to the degree of transparency chosen by a central bank. Second, virtually all other studies, including Thornton (1996), Muller and Zellmer (1999), Rafferty and Tomljanovich (2002), and Coppel and Connolly (2003) analyze a single central bank. Instead, we directly compare central bank policies and financial effects in seven developed countries as a means of controlling for changing national and global conditions across the time span. This cross-country methodology also allows for a comparison between central banks that have moved towards greater disclosure and those central banks that have remained opaque. Third, we attempt to disentangle the effects of changing transparency from other central bank procedural changes, such as independence from the national government or implementing an inflation targeting regime.

The paper proceeds as follows. Section 2 lays out the definition of central bank transparency and the economic and political issues surrounding transparency, and summarizes the literature. Section 3 describes the data and models used. Section 4 outlines the main results. Finally, section 5 offers policy recommendations and concludes.

2. Background and Related Literature

Defining Transparency

Complete transparency simply means that the central bank and the public have access to the same information, incomplete or uncertain though it may be, when making economic decisions. (3) Since the central bank generally has access to information first, transparency then suggests that the central bank "accurately" passes information on to other agents. (4) "Accurately" is difficult to define in practical terms. Though many economists believe in the "more is better" creed, a distinction needs to be made here between accuracy and clarity due to the sheer amount of quantitative and qualitative information a central bank collects in the process of policy implementation. One possibility open to a central bank in a move toward greater transparency is to flood the public with all such internally gathered information (such as, for example, on the central bank's website). The obvious drawback to this approach is that the time and expertise needed by members of the public to sift through the information renders timely decision-making nearly impossible. Therefore, greater transparency by this benchmark may in fact hinder economic decisions and thus reduce social welfare.