Journal of International Money and Finance (October 2023 issue, Vol. 137)

Working paper version:
Available on SSRN (paper version June 2022)

Abstract: Does communication influence trust in the central bank? We examine this question using survey data covering 488,000 Eurozone citizens from 1999 to 2019. We find compelling evidence that more communication, as measured by the number of speeches made by Eurosystem representatives, negatively impacts citizens’ trust in the ECB. This holds for speeches as a whole and for different groups of speakers. This effect was exacerbated during the global financial and European sovereign debt crises. We do detect a positive result from more speeches in the form of increased informedness on the ECB and the EU. However, the overall negative effect prevails.

Media coverage of this paper:

• "You should speak less often, Ms. Lagarde" ("Halten Sie öfter den Mund, Frau Lagarde"; Frankfurter Allgemeine FAZ, 05-Aug-2022).
• "Some unpleasant speaking arithmetic: ECB speeches and trust" (Vox, CEPR's Policy Portal, 11-Jul-2022).

Journal of Money, Credit and Banking (October 2022 issue, Vol. 54, No. 7, pp. 2125 — 2148)

Working paper version:
WWZ Working Paper 2018-05 (paper version January 2018)
SNB Working Paper 2017-17 (paper version November 2017)

Abstract: We demonstrate how the precision of public and private information can be measured and their welfare implications can be assessed. To this end, we develop and apply a procedure to test welfare results from a beauty and nonbeauty contest model based on survey forecasts of interest rates in a large country sample over an extended period of time. In most countries, interest rate forecasts are unbiased. In half of the countries, a higher precision of public information regarding interest rates increases welfare. This applies in particular to small open and emerging market economies. During forward guidance, public information is less precise than private information. On average, forecasters place greater weight on public than private information and rapidly correct errors contained in private information, while errors in public information last over the whole forecast horizon.

Media coverage of this paper:

• "More frequent central bank communication worsens financial and macroeconomic forecasts" (Vox, CEPR's Policy Portal, 04-Mar-2018).

Journal of International Money and Finance (July 2021 issue, Vol. 115)

Working paper version:
SNB Working Paper 2020-17 (paper version July 2020)

Abstract: We analyze the economic impact of central banks sensed by business executives in a sample of 61 countries from 1998 to 2016. Based on a survey conducted by the Institute for Management Development (IMD), we find compelling evidence that intensive central bank communication, as measured by the quantity of speeches, worsens the perceived impact. During the global financial crisis (GFC), this effect became even stronger. In contrast, economic growth and a positive output gap improve the opinion executives have of their central bank's impact on the economy. Moreover, although less robustly, higher unemployment, and higher short-term interest rates worsen executives' opinion, while market uncertainty improves it. The level of inflation and an inflation targeting regime, central bank independence and transparency, financial crises, the zero lower bound constraint, forward guidance, the performance of the stock market, and the volatility of the exchange rate seem to be unimportant in this regard. A more detailed analysis shows that the effect of speeches depends on the identity of the speaker, the exchange-rate regime or an explicit employment mandate.

Media coverage of this paper:

• "Central bank communication: Remember who's talking" ("Zentralbank-Kommunikation: Es ist entscheidend, wer spricht"; Ökonomenstimme, 24-June-2021).
• "Central bank communication: Remember who's talking" (Vox, CEPR's Policy Portal, 12-May-2021).
• "Should Thomas Jordan blog and twitter?" ("Sollte Thomas Jordan bloggen und twittern?"; Sonntagszeitung, 13-Sept-2020).

International Journal of Central Banking (March 2020 issue, Vol. 16, No. 2, pp. 153 — 201)

Working paper versions:
WWZ Working Paper 2018-06 (paper version January 2018)
SNB Working Paper 2017-12 (paper version October 2017)

Abstract: 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.

Media coverage of this paper:

• "Two Audiences, One Lectern" (by Peter Coy, TECH ICONS, 10-Jun-2025).
• "Let's Not Return to the Bad Old Days of Central Bank Secrecy" (Bloomberg View, 14-Mar-2018).
• "A Cacophony of Central Bankers" (Wall Street Journal, 05-Mar-2018).
• "More frequent central bank communication worsens financial and macroeconomic forecasts" (Vox, CEPR's Policy Portal, 04-Mar-2018).
• "When central banks inform too much" ("Wenn Notenbanken zu viel informieren"; Zentralschweiz am Sonntag, 05-Nov-2017).
• "On forward guidance" (e-axes 360° Econ View, 12-Oct-2017).
• "Central Bankers Fare Better When They Are Tight Lipped" (Bloomberg, 11-Oct-2017).
• "More communication may hamper monetary policy, paper says" (Central Banking, 09-Oct-2017).

The dataset for our communication measure is available upon request, see section Data.

In Expectations: Theory and Applications from Historical Perspectives
(March 2020, Springer Studies in the History of Economic Thought, Arie Arnon, Warren Young, Karine van der Beek (eds), Springer, pp. 215 — 238)

Working paper version
CEPR Discussion Paper DP13187 (paper version September 2018)

Abstract: Even when all past and present information is known individuals usually remain uncertain about the permanence of observed variables. After reviewing the history and role of adaptive expectations and its statistical foundations in modeling this permanent-transitory confusion the paper investigates the consequences of this confusion for tests of market efficiency in the treasury bill and foreign exchange markets. A central result is that the detection of serial correlation in efficiency tests based on finite samples does not necessarily imply that markets are inefficient. The second part of the paper utilizes data on Israeli inflation expectations from the capital market to estimate the implicit speed of learning about changes in inflation and to examine the performance of adaptive expectations in tracking the evolution of those expectations during the 1985 Israeli shock stabilization as well as during the stable inflation targeting period.

Media coverage of this paper:

• "The Permanent-Transitory Confusion: Implications for Tests of Market Efficiency and for Expected Inflation During Turbulent and Tranquil Times" (Vox, CEPR's Policy Portal, 04-Nov-2018).

Available on SSRN (paper version February 2025)

Abstract: We examine whether and how different types of Eurosystem speeches affect citizens' support for the euro. Using country-level panel data analysis and individual-level regression analysis for 19 euro-area countries over the period 1999-2023, we find compelling evidence that speeches delivered by representatives of national central banks significantly raise people's support for the euro. By contrast, speeches by Executive Board members do not affect public perception of the single currency.

Media coverage of this paper:

• "Why trust in the ECB and support for the Euro are not the same thing" (EUROPP European Politics and Policy, LSE Blog, 23-Jul-2025).
• "Central bankers’ speeches good for the euro, but not ECB – study" (Central Banking, 23-Jul-2025).

CEPR Discussion Paper DP15039 (paper version July 2020)

Abstract: We examine the cross-country relationships between measures of forecast uncertainty, forecast dispersion across individual forecasters and the variabilities of short-term interest rates and long-term yields. The main findings are: (i) Forecast uncertainty and forecast dispersion are positively and significantly related across countries for both short-term interest rates and long-term yields. (ii) A positive, albeit weaker, relation is found between forecast uncertainty and interest rate variability. (iii) Forecast dispersion of short-term interest rates and rates' variability are also positively associated. The evidence is followed by a Bayesian learning model that discusses conditions under which the results above are implied by theory.

SNB Working Paper 2018-10 (paper version May 2018)
CEPR Discussion Paper DP12489 (paper version December 2017)

Prepared for the 4th Thomas Guggenheim Conference on "Expectations: Theory and Applications in Historical Perspective", at Ben-Gurion University of the Negev, December 2017.

Abstract: This paper develops a model of honest rational professional forecasters with different abilities and submits it to empirical verification using data on 3- and 12-months ahead forecasts of short-term interest rates and of long-term bond yields for up to 33 countries collected by Consensus Economics. The main finding is that in many countries, less-precise forecasters weigh public information more heavily than more-precise forecasters who weigh their own private information relatively more heavily. One implication of this result is that less-precise forecasters herd after more-precise forecasters even in the absence of strategic considerations. We also document differences between the average forecasting errors of more- and less-able forecasters as well as substantial correlations between the forecast errors of different forecasters.

SNB Working Paper 2018-14 (paper version November 2018)

Abstract: In this paper, I derive and apply three univariate methods and one bivariate method to estimate permanent and transitory components of the American output growth path during the 1790 to 2017 period. The results show that statistical tests give little support to the hypothesis of significant permanent growth rate changes (univariate methods). The "special century" (1870-1970, as defined by Gordon (2016)) exhibited more volatile permanent shifts in the output level compared to recent decades (bivariate method).