What the Hawkometer measures

The Hawkometer answers one question: based on what a senior central bank official has said in their public appearances, where do they sit on the hawkish-dovish spectrum, and is that position drifting?

It is not a measure of:

• What rate they will vote for at the next meeting (that is what our probability and Taylor Rule tools are for).
• Their long-run reaction function.
• Their personal beliefs separate from the committee’s published guidance.

It is a measure of public communication tone, normalised across speakers and aggregated up to the committee level.

The scoring engine

Every speech, interview, testimony or press conference is run through a phrase scorer:

score(text) = clip(-10, +10,
                   sum over hawkish phrases of (weight * f(count))
                 - sum over dovish phrases of (weight * f(count)))

Where f(count) applies sub-linear scaling for repeated phrases (f(1) = 1, f(2) ≈ 1.69, f(3) ≈ 2.10). This prevents a speaker from gaming the score by repeating one phrase ten times, while still rewarding emphasis when a phrase is used more than once.

The clipping at +/- 10 ensures that no single speech, no matter how strongly worded, can dominate a committee aggregate.

The phrase library

The library is in two parts. Hawkish phrases push the score up; dovish phrases push it down. A representative sample is shown below; the full library is published as part of the data feed (hawkometer.json → phrase_library).

Hawkish phrases (selection)

Dovish phrases (selection)

Some phrases are deliberately ambiguous — “data-dependent” is technically a process statement, but in current usage it skews mildly dovish, so it sits at -0.7. We document each judgement call rather than hide it.

From speeches to committee scores

Once individual speeches are scored, we aggregate in three layers:

1. Per-speaker rolling 90-day average

For each official we compute the simple average of their sentiment scores over the last 90 days. We also compute a 30-day average and a prior-60-day average so a shift indicator can be reported (last 30 days vs. the 60 days before that).

2. Per-bank, voter-weighted committee score

The committee score is the weighted average of each speaker’s 90-day score:

A non-voter is still part of the committee’s intellectual centre of gravity, so we don’t drop them entirely — but a voter’s voice counts roughly twice as much in the index, which matches how markets price these officials.

3. Cross-bank lean

Per-bank scores are translated into a verbal lean for human readers:

These are descriptive labels, not predictions about the next decision.

Limitations — read this before citing the index

The Hawkometer is a useful first cut, but it has real limitations. Anyone citing it in research or reporting should understand them.

Context blindness. The scorer matches phrases, not arguments. A speaker saying “the case for additional tightening has weakened materially” will, in the current implementation, register additional tightening as a hawkish phrase even though the surrounding sentence makes the meaning dovish. We mitigate this by aggregating across many phrases and many speeches; we have not yet attempted to handle negation explicitly. This is a known weakness.

Novel language. When an official invents a new turn of phrase that markets immediately interpret as hawkish or dovish, the index will not pick it up until we add the phrase to the library. We do a quarterly review and an emergency update after major regime-change speeches.

Hypotheticals and conditionals. “If inflation were to re-accelerate, additional tightening would be appropriate” reads more hawkish to the scorer than the speaker intended. Again, weight comes from many speeches, not from a single sentence — but readers should not over-interpret a single high reading from a single appearance.

Translation effects. ECB speakers, BoJ officials and SNB Board members frequently speak in languages other than English. Our scorer currently runs only on the English version of those remarks. When the official version is non-English, scoring is delayed until a translated transcript is published; for press conferences we use the simultaneous interpretation transcript. This introduces a small lag and a small translation bias.

No causal claim. The index does not claim that a hawkish reading causes anything. It is an organised summary of what officials have said. Use it alongside market-implied probabilities, the Taylor Rule analysis and the rate path tools.

Sample data. While the production scrapers ramp up, some entries on the site may be representative samples rather than direct transcripts. Each sample-derived speech is flagged with is_sample: true in the JSON feed so downstream consumers can filter them out if needed.

Update cadence

The pipeline runs daily on the Central Bank Watch backend. New speeches typically appear within a few hours of delivery once the production speech feeds are connected. The phrase library is reviewed at the end of each quarter and any time a major central bank materially changes its communication style.

Reproducibility

Every score is fully reproducible from the raw text and the phrase library. The complete dataset (speakers, speeches, aggregates and the full phrase library) is published as hawkometer.json in the site’s data feed. If you find a scoring decision you disagree with, you can reproduce the calculation by hand and tell us which phrase you would weight differently.

For methodology questions or to suggest a phrase to add or re-weight, see the about page.