How to Fix the ECB’s Communication Problem

The ECB has a serious communication problem. It has undertaken a number of unconventional measures lately–the three year LTROs, the acceptance of questionable assets for collateral, and the large expansion of its balance sheet–that amount to a QE program by stealth. And therein lies the problem. The ECB’s actions have not been communicated to the public ex-ante nor have they been linked to a targeted outcome. In short, the ECB is failing to manage expectations, the most important monetary policy transmission channel at its disposal.

Now Goldman Sachs would say that the above is a charitable interpretation of the ECB’s recent actions. They note in a research paper (via Cardiff Garcia) that the ECB is not just failing to manage expectations, but is actually destabilizing expectations.

The ECB offers little ex ante information about its outright asset purchases via the securities markets programme (SMP). The stock of outstanding purchases is only revealed ex post, and no information ispublished on the composition of that stock, either by maturity or by country of issuer. There is no preannounced schedule of purchases. Market participants thus face substantial uncertainty about when and where the ECB will intervene: this probably serves to reduce the liquidity of the underlying market and precludes the possibility that the market will anticipate the ECB’s actions, helping policy makers to achieve their policy objectives.

This approach extends to other aspects of the ECB’s nonstandard policy measures. For example, the ECB has yet to announce whether further 3-year LTROs will be conducted beyond February this year. Such an announcement would serve to help stabilise market conditions further, by providing further reassurance about the availability of funding over longer horizons. Should we not see such an announcement at the February Governing Council meeting, the potential for a regression in market developments is obvious. The kink in peripheral yield curves at around 3-year maturities is evidence of this concern.

More generally, the wider ECB communication surrounding its enhanced credit support has always been grudging: the ECB has, at times, appeared reluctant to offer such support, despite the fact that, in practice, it has done so in vast amounts. By implication, the reassurance offered to households and firms about the ECB’s commitment to macro stabilisation has been weakened, to the prejudice of the effectiveness of the policy as a stabilisation tool.

There is an easy solution to the ECB’s communication problem that directly and aggressively addresses the insufficient aggregate demand problem while still maintaining a long-run nominal anchor. And to boot, it does not depend on bank lending (though financial intermediation would probably increase as a result). The solution is setting a nominal GDP level target. The level part is important because it signals clearly to the public that the ECB would commit to buying up (or selling) as many assets as needed until nominal GDP hit some pre-crisis trend path. Not only would this fix many of the Euro debt problems, it would create more certainty and cause the public to much of the heavy lifting in restoring aggregate demand (i.e. the public would adjust their portfolios in anticipation of the ECB buying up more assets and in the process cause nominal spending to adjust largely on its own. See here for more details.) This is what makes the ECB’s floundering so frustrating for me to watch.

Disclaimer: This page contains affiliate links. If you choose to make a purchase after clicking a link, we may receive a commission at no additional cost to you. Thank you for your support!

About David Beckworth 240 Articles

Affiliation: Texas State University

David Beckworth is an assistant professor of economics at Texas State University in San Marcos, Texas.

Visit: Macro and Other Market Musings

Be the first to comment

Leave a Reply

Your email address will not be published.


This site uses Akismet to reduce spam. Learn how your comment data is processed.