Do private forecasters in Chile take into account the Central Bank’s projections when updating their expectations?

Keyword: 
Monetary policy
Topic: 
Macroeconomics - Economic growth - Monetary Policy

Expectations about the future economic development are important for the private agents’ economic decisions, and knowledge about how these expectations are formed helps monetary policy makers perform a more efficient conduct of the policy. Hence, a perception of which factors affect private expectations and, particularly, whether the Central Bank can influence them is an important issue. A recent paper by Pedersen (2013) applies Chilean data to analyze which factors affect the private forecasters’ expectations about the future inflation and growth rates and, particularly, if the forecasts published by the Central Bank of Chile (CBC) influence these expectations.

The CBC publishes its forecasts in the Monetary Policy Report (MPR) every quarter, and it also conducts a monthly survey amongst private forecasters (PFs) — the Economic Expectations Survey (EES) — to reveal their expectations about the future values of certain economic variables. To investigate if the PFs react to the predictions published by the CBC, their expectations are compared before and after the publication of the CBC forecasts such that the focus is on how private forecasters (PFs) update their expectations. It might be expected that PFs adjust their predictions towards those of the CBC only if they believe that the CBC’s forecasts are better than their own, maybe because they think that the CBC holds valuable information that is not publicly obtainable or that it has more resources available for making the forecasts.

Available data allow for an analysis of short- and medium-term inflation expectations (end of current and next year, respectively) and short-term growth expectations. A plot of the update of these expectations against the difference with respect to the CBC forecast can give a first impression of whether PFs take into account the predictions of the Central Bank when updating their expectations. Figures 1 to 3 show these plots and it seems that the PFs’ short-run inflation expectations do take into account the predictions of the CBC, while it is not evident for the medium-term expectations and it is hard to see any relationship between the growth rate update and the forecast of the CBC.

Of course there may be several other reasons, unrelated to the CBC forecasts, for the PFs to change their expectations. The most relevant ones seem to be news in the form of surprises in the monthly published observation and changes in the expectations regarding the exchange rate, the monetary policy rate and the oil price. On the other hand, disagreement amongst the forecasters does not matter for the formation of the expectations and neither do published revisions of the growth rate or changes in the forecasted US growth rate.

Inflation forecasts - current year

Inflation forecasts - next year

Growth forecast - current year

It would be reasonable to believe that PFs take into consideration the CBC’s forecasts only to the extent that they think the Central Bank is credible. Measuring Central Bank credibility is not easy, but a simple measure, for which data are available in Chile, is if the PFs believe that the Central Bank will reach its target. The target of the CBC is that inflation should be at 3 per cent in a horizon of two years, which is also one of the questions in the EES. Hence, a simple measure of Central Bank credibility is the divergence between this measure and 3 per cent. It turns out, however, that there is no strong evidence that PFs include their assessment of the credibility when updating their expectations about future growth and inflation.

An econometric analysis including the variables discussed earlier suggests that indeed the PFs do take into account the CBC forecast when updating their short-term inflation expectations. PFs also consider surprises in the monthly published inflation data as well as changes in the expectations about the future exchange rate and monetary policy rate. On the other hand, it is not evident that medium-term inflation expectations are directly influenced by the CBC’s forecast, but indirectly they are as a main factor explaining these updates is changes in the PFs’ short-run expectations. Also changes in the expected monetary policy rate and oil price are important when PFs update their medium-term inflation expectations. The growth rate expectations are not influenced by the projections of the CBC. These are affected by surprises in the monthly published data and changes in the outlook for the monetary policy rate.

With respect to the influence of the CBC’s projections, a couple of natural questions arise: (1) Does the timing of the publications of the CBC’s forecasts matter, i.e. do PFs take into consideration differently when the forecasts are published in the beginning of the year and in the end of the year? (2) Does it matter if the difference between the PFs’ and the CBC’s forecasts are negative or positive? The answers to these questions are “yes” and “yes”. The short-run inflation expectations of the PFs seem to be affected by the CBC’s forecasts mainly if they are published in the first half of the year and when they are lower than the PFs expectations. Medium-term expectations are influenced directly by the CBC forecast published in the last quarter and if it is lower. Finally, independently of the timing of the publication and whether the CBC’s forecast is higher or lower, PFs’ growth expectations do not seem to be affected by the forecasts of the Central Bank.

A tempting recommendation from these results would be to promote that central banks publish their forecasts, especially those for inflation, to help PFs in the formation of their expectations and thereby facilitate the conduct of the monetary policy. The important role of the monetary policy rate expectations in forming the private predictions suggests that special attention should be paid to the development of the expectations about the future policy rate.


References

Pedersen, M. (2013), ‘What Affects the Predictions of Private Forecasters? The Role of Central Bank Forecasts’, Working Paper No. 686, Central Bank of Chile.

Disclaimer: The views in and conclusions expressed in this blog do not necessarily represent those of the Central Bank of Chile or its Board members.
 

 

 

Share this