What we learned covering a rate cycle

The economics desk has now covered a full interest-rate cycle: the rapid tightening that began after the inflation surge, the long pause at the peak and the first steps of easing. Looking back at what we wrote over that period is uncomfortable in places. Here is what we would do differently.

Forecasts were less useful than we treated them

We reported the market-implied path for rates, and the consensus of economists, as if they were information about the future. They were information about the present: what participants believed at that moment. The market priced a peak rate that was too low for most of the tightening phase, then priced cuts that arrived a year later than expected. The economists' consensus was not better.

We now report those forecasts with their track record attached. When we write that markets price a cut, we note how often the pricing three months ahead has matched what happened. Readers deserve to know that the number is a mood, not a schedule.

The language of central banks is designed to be misread

Policymakers choose words carefully, and we spent too long reading their statements as if they were precise. "Data-dependent" was used to mean both "we will cut if inflation falls" and "we will not commit to anything". "Restrictive" described a stance that policymakers themselves disagreed about. The lesson was to quote the words, describe the range of interpretations and resist choosing one on the central bank's behalf.

Lags are real, and they are longer than the news cycle

The effect of a rate change on the economy takes a year or more to arrive. The news cycle runs on a day. For much of the tightening phase we wrote about the economy's resilience as if it disproved the effect of higher rates, when it mostly reflected loans that had not yet repriced. We now try to say where in the transmission process the economy sits, and to point to the indicators that lead rather than the ones that lag.

The consensus can be wrong in the same direction for a long time

Every quarter for two years, the majority view was that a recession was months away. It did not arrive. A newsroom that reports the consensus faithfully will be wrong alongside it, and our readers will not thank us for company. We now make a point of finding and presenting the strongest minority case, not for balance, but because the minority has been right often enough to earn the space.

What held up

Reading the primary documents, comparing statements line by line and computing our own figures all proved their worth. The stories that aged best were those that described what had happened and what the data showed, and were modest about what came next. The ones that aged worst told readers what the central bank would do.

The next cycle will be different in its particulars and the same in its lessons. We intend to remember them.