Collections > UNC Chapel Hill Undergraduate Honors Theses Collection > A Measure of Monetary Policy Transparency: How Newsworthy are the Actions of the FOMC?
pdf

There are two main goals of this paper. The first goal for this paper is to create a quantifiable measure for transparency that captures public interpretation of FOMC policy decisions. We focus on the extent to which the FOMC conveys itself and its policy decisions with their official statements, which are released immediately after each meeting, to the public at large. To do this, we create an index that attempts to isolate the communicative success between the FOMC and the public in a twofold process that involves a trio of documents. First, we need to account for how the Federal Reserve’s practice of transparency has evolved over time. Namely, we need to be sensitive toward (i) the Fed's institutional changes in its release of materials and (ii) the differing lag times at which these materials were released. To do this, we first establish transparency by computing the similarity values between discourse within the transcripts and reporting in the statements (previously the Record of Policy Actions) are high. This step is necessary as it not only establishes the Fed's definition of central bank transparency, but it also acts as a proof of concept in preparation for our second step. Next, we compute the similarity between the statements and the news articles that report on the statements. If we want to provide meaning to this definition of transparency, understanding public interpretation of Fed policy is necessary. I create a transparency index series from 1984 to 2012 using a computational linguistics technique called Latent Dirichlet Allocation (LDA), and the idea is that a higher similarity number indicates that the FOMC is more transparent in their communication. My second goal is to understand if transparency affects the way monetary policy is propagated.