Central banking can be agonising. The effect of monetary policy on the economy is not immediate,
so decisions must be based on expectations for two years' time. That means putting faith in forecasts that could well turn out to be wrong.
Some soul-searching might be expected at the monetary-policy meeting of the European Central Bank (ECB) on December 13th.
Since 2015 it has stimulated the euro-zone economy by buying bonds worth 2.6trn euros ($3trn) under its quantitative-easing programme.
In June it said it planned to stop asset purchases by the end of the year (though it will continue to reinvest proceeds from maturing bonds),
and to keep interest rates at their current rock-bottom levels "through the summer" of 2019.
Since then, though, economic growth has slowed and inflation, stripped of oil-price rises, has stayed resolutely low.
The question is how persistent these developments will be.
After an impressive showing in 2017, the euro zone's economy has weakened. Temporary factors bear some of the blame.
Poor weather, strikes and a bad flu season marred the start of the year.
Then in the third quarter of 2018 output was 0.2% higher than in the second, the slowest growth since 2014.
New emissions tests in September meant carmakers put production on hold while they rushed to certify their stock of vehicles.
As a result, in the third quarter the German economy— the bloc's powerhouse—contracted for the first time in nearly four years.
In 2017 trade provided a welcome boost to the euro zone.
This year demand for exports has moderated, dampening growth, most notably in Germany and Italy.
Domestic demand has picked up some of the slack, particularly in France and Spain. But it has not fully offset the effects of the trade slowdown on the bloc.