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

Southern Economic Journal, Jan, 2007 by Marc Tomljanovich

What effect does increased central bank openness have on the predictability of short-term rates? Tabellini (1987) models secrecy as generating parameter uncertainty. Through a Bayesian rule, agents have the opportunity to learn about the central bank's secret policy target. However, this learning process leads to greater volatility in asset prices, as the market initially overreacts to changes in monetary policy changes. Dismantling Dotsey's (1987) and Rudin's (1988) claims to the contrary, Tabellini (1987) proposes that secrecy tends to increase the unconditional variance of the Federal Funds rate. In addition, opacity increases the conditional variance of the forecast error of the funds rate. Finally, Tabellini (1987) explains that this volatility carries over to market interest rates of longer maturities.

Taken together, Mankiw and Miron (1986) and Tabellini (1987) show how increased central bank transparency leads to less market uncertainty about parameter estimates, helping to form more accurate interest rate forecasts and hence greater market predictability. These more precise forecasts increase plim [beta], making it more likely that the expectations hypothesis holds. In addition, if their assumptions hold, the expectations theory in general and Equation 3 in particular should more closely fit the data after the move to greater openness.

Empirical Model

We now turn to a more general version of the expectations theory of the term structure. According to the simple expectations hypothesis, [beta] = 1 in the following regression equation

[S.sup.(n,m)*.sub.t] = [alpha] [S.sup.(n,m).sub.t] [[epsilon].sub.t], (5)

where [S.sup.(n,m).sub.t] = [R.sup.n.sub.t] - [R.sup.m.sub.t], [[epsilon].sub.t] is an error term, n is a constant, n is the duration of the long-term bond, and m is the duration of the short-term bond. Accordingly, [R.sup.n.sub.t] is the interest rate on the long-term bond at time t and [R.sup.m.sub.t] is the interest rate on the short-term bond. [S.sup.(n,m)*.sub.t] is the perfect foresight spread, which is defined as

[S.sup.(n,m)*.sub.t] = [k-1.summation over (i=1)] (1 - i/k) [[DELTA].sup.m][R.sup.m.sub.t im], (6)

where k = n/m and [[DELTA].sup.m] means the change is calculated over m periods. (14)

Estimating Equation 5 with data before and after the policy change allows us to determine if there has been any improvement in the performance of the expectations hypothesis since the policy change. If increased openness improves the efficiency and reduces the volatility of financial markets, as claimed in the previous subsection, then one would expect the expectations hypothesis to perform better after a central bank's move to greater transparency. An improvement of the model may take the form of a larger slope coefficient and an increase in the explanatory power of the expectations hypothesis.

To test this hypothesis, we calculate the Root Mean Squared Error (RMSE) of the forecast for each interest rate pair both before and after the central bank's policy change. If increased openness increases efficiency and decreases volatility, then one would expect the RMSE to decrease after the policy change. If the increased transparency provided new and relevant information to markets, then one would expect that the errors caused by incorrect guesses regarding central bank monetary policy changes would become less frequent. This would eliminate one source of error in the market and reduce the RMSE of forecasts based upon market information, ceteris parabis.