Do central bank speeches effect macroeconomic forecasts?

Interesting paper by SNB economists – Thomas Lustenberger and Enzo Ross.

The general belief is that more communications and transparency is good and more should be encouraged.  However, the authors of the paper don’t think so:

In a large sample of countries across different geographic regions and over a long period of time, we find limited country- and variable-specific effects of central bank transparency on forecast accuracy and their dispersion among a large set of professional forecasts of financial and macroeconomic variables. More communication even increases forecast errors and dispersion.

This will surely please current management  of Indian central bank where communications has been reduced to bare minimum.

More interestingly, they look at multiple variables to assess communications and transparency. One of them is central bank speeches. Guess which central bank has given most speeches  from  1998-2014? Federal Reserve of course. Guess which is second? RBI!

We broaden the coverage of actual central bank communication by constructing a comprehensive and explicit measure of communication consisting of central bank speeches. To this end, we compiled a variable made up of central bank speeches as collected by the Bank for International Settlements (BIS). For each central bank reporting their speeches to the BIS, we counted the number given in the month preceding the forecast. For this variable, we have observations from 1998 to 2014.

Table 1 exhibits the total number of speeches per country and their monthly average divided by four geographic areas. As can be seen, most speeches are given by central banks in Western countries (WE), above all by the Federal Reserve (1,386) and Japan (453). Indian central bankers, grouped with the Asia-Pacific countries (AP), delivered the second-highest number of speeches (648).

Figure 1 illustrates how communication activities by central banks have intensified over time. The number of speeches has steadily increased from approximately 150 in 1998 to nearly 900 in 2013 and 2014. Importantly, this not only reflects more communication activities but also a
higher number of central banks reporting their speeches to the BIS.

An important issue is the potential endogeneity of communication. The number of speeches could be endogenous to the economic situation. In a more uncertain environment, forecasts are likely to be both more inaccurate and more disperse. For this reason, the central bank may want to increase its communication activities. We cannot exclude that some of the speeches were the result of unexpected events that the central bank considered important enough to justify intervention. However, for the bulk of the speeches, this is a very unlikely outcome. Speeches by central banks are fixed and announced months in advance, making our communication proxy a well-defined exogenous variable.

They also find that these speeches create a bigger impact on forecasts:

The main and most important result of our analysis relates to communication, which, as discussed, is measured by the number of central bank speeches. It exerts a much greater influence on private forecast performance than transparency. From Table 3 and Table 4, we can deduce that intensive communication activities make it more difficult to forecast inflation, yields (the latter in contrast to what was found for transparency) and GDP growth, and they increase the dispersion in forecasts of inflation, yields and short-term rates. In terms of statistical significance, the effect of speeches on inflation forecasts is highest. We provide a discussion and interpretation of these results in Section 6.

Lots of econometrics etc in the paper. Need to read carefully (something this blog keeps saying but never manages to do!)….