Hawkish vs Dovish: It Is the Lean, Not the Rate

By Uladzimir Leuchankou, founder of Signal2Forex

What is a hawkish central bank, and what does dovish mean? The usual answer to the hawkish vs dovish question is that a hawkish bank raises rates and a dovish bank cuts them. I suppose that is an answer too. But for me the two words have never been about the rate itself. They are about the lean — which way the bank is going to move next, and how badly it wants to get there. A bank can sit at 1.25% and be the most hawkish institution in the world. The Bank of Japan is doing exactly that this week.

I run trading robots on fourteen currency pairs, on a real account that is streamed on our YouTube channel. Every central bank decision goes through that account whether I like it or not. So I have had to learn these two words the practical way, and this week, with four big banks deciding inside nine days, is a good moment to write down what I actually look at.

Hawkish vs dovish: two words, one lean

Hawkish meaning, in one line — a bank that worries about inflation more than about growth, and so wants higher rates or at least no cuts. Dovish meaning — a bank that worries about jobs and growth more than about prices, and so wants lower rates or at least no rises. That is the dictionary meaning of hawkish and dovish. The opposite of hawkish is dovish, and a bank sitting between the two is called neutral. Hawks attack, doves wait.

For a currency trader the hawkish vs dovish meaning is narrower, and what is hawkish in forex is not quite what is hawkish in the news. A hawkish stance is good for the currency, because a higher expected yield pulls money in. A dovish monetary policy is bad for it, because the money drifts out to wherever the yield is better. Not only the current rate matters here, but the whole path of rates the market expects for the next year. That path is what the pair actually prices.

There is one more thing, and it is the thing that cost me money when I started. The market has already made its guess before the meeting. If the bank does what the market guessed, the currency hardly moves — even when the headline sounds dramatic. If the bank is more hawkish than the guess, the currency rises. If it is more dovish, it falls. Poorly read, a rate rise can send a currency down. Well read, a hold can send it up.

Nine days in October

Here is the week as it stands on Friday 30 October 2026, bank by bank, with my reading of whether each one came out hawkish or dovish. Three have decided, one is still to come.

Four decisions, three already made. Not one of them changed a rate this week, and every one of them leaned the same way.

The Fed on Wednesday is the cleanest lesson. The rate stayed at 3.75–4.00%, which roughly nine traders in ten expected. Then you read down to the vote: 10 to 2, with Lisa Cook and Beth Hammack voting to raise now. One word in the statement changed too — the economy went from “solid” in September to “strong”. Chair Kevin Warsh called financial conditions “tighter, but not tight” and would not say what December brings. So is the Fed hawkish or dovish right now? Hawkish, with the rate standing still. Nothing moved, and the lean moved a lot.

Futures now put a December rise at roughly 70–80%. That number, not the rate, is what Wednesday changed.

The ECB on Thursday is the harder one, and it is the kind of meeting these two words were invented to argue about. The written statement was softer than September’s: more about the energy shock not spreading, more about tighter credit slowing growth. Reading only the text, I would have called it dovish.

Then Lagarde spoke. She said a rise had been discussed, that December would bring the data to decide, that “the direction is quite clear”. She also said she will leave on 7 February 2027, nine months early. The euro fell on the day. For me the honest label is two labels — a dovish text with a hawkish tone in the press conference — and when the two disagree, I have learned to trust the tone of the voice over the text.

The Bank of Japan this morning did what we expected on Wednesday, and I was still glad to see it. The rate stayed at 1.25%, the highest since 1995. Hajime Takata voted against the hold, in favour of a rise. The new Outlook Report raised the inflation forecast and the bank repeated that it is ready to keep going. Rate unchanged, lean upward, one dissent — the same shape as the Fed two days earlier.

Votes before words

Is there one signal I would keep if I had to throw the rest away? Yes, and it is not the statement. It is the vote count. A statement is written by a committee and every sharp word in it gets sanded down before release. A vote cannot be sanded. When one member votes to raise, the hawkish side has grown by one. When the chair lets two such votes through instead of talking them out of it, the whole room has moved.

The Bank of England this year is the example I show people. In February four of nine members voted to cut. By June two voted to raise. By July, three — Greene, Mann and Pill — and the same three again in September. Bank Rate did not change at any of those meetings. Someone watching only the rate saw six months of nothing. Someone counting the votes saw the committee turn through 180 degrees, and the pound priced the November rise long before the Bank announced it.

The Fed did the same in miniature: three votes to raise in July, no change, then the rise itself in September. Nobody who had counted in July was surprised.

I did not count votes when I started — I read the rate, decided the meeting was a non-event and went back to the charts. The statement had the dissents in it the whole time. If you take one habit from this page, take that one.

My three checks

Over the years the routine before a decision has shrunk to three checks. None of them is the expected rate, because the expected rate is already in the price.

The first check — what the market expects, in numbers. Not “a hold is expected” but the probability from rate futures, and the expected rate three and six months out. On Wednesday a Fed hold was about 90% expected and a December rise about 70–80%. Those two numbers are the ruler. I measure the statement against them, not against the previous statement in the abstract.

The second check — the old statement next to the new one. Central banks change as few words as they can, so every change is on purpose. “Solid” becoming “strong” is a hawkish edit. A sentence about “today’s action” disappearing because there was no action is just housekeeping. It takes a few meetings to tell the two apart, and after that you cannot unsee it.

The third check — the vote, and the names. Two dissents from the same side matter. Two from opposite sides cancel. A governor matters more than a regional president, because governors vote at every meeting.

And one thing I do not do: trade the first five minutes. Spreads on the major pairs get several times wider at the release, and the first move often turns around when the press conference starts. My robots have a spread filter for this. When the spread widens past the threshold, they stop opening new orders and wait.

On Wednesday the Forex V Portfolio on the live account sat through the Fed announcement with no new orders and picked up again about twenty minutes later. You can do the same by hand: switch AutoTrading off, make tea, come back.

Mistakes that cost money

Why do beginners get these two words wrong so often? Mostly because they stop at the dictionary. Four mistakes I see every month.

Hawkish is not the same as bullish, and the hawkish vs dovish vs bullish vs bearish muddle costs beginners real money. Hawkish is good for the currency, bad for gold, bad for long bonds, usually bad for shares. The word describes the bank. Bullish describes a price, and you have to say which price.

A hold is not neutral. The Fed held on Wednesday and the ECB held on Thursday. One was a hawkish hold with two votes to prove it. The other was a soft text with a hawkish press conference. Same decision on paper, different lean, different move in the pair.

A rise is not always hawkish. The Bank of Japan raised rates on 18 September and the yen fell more than 1% the same day — USD/JPY went from about 155.5 to above 157 within hours. The rise was fully expected. The two votes against it and the careful guidance were not. The surprise was dovish, and it was hiding inside a rise.

The other currency has a bank too. A hawkish Fed against a hawkish Bank of England is a wash. A hawkish Fed against an ECB that just softened its text is a trade. I never read one bank alone — I read the gap between two.

What comes after November

Next Thursday, 5 November at 12:00 GMT, the Bank of England decides. The quarter-point rise to 4.00% is priced above 90%, so the rise itself is not the news. The vote split is — 5 to 4 or 8 to 1 are very different messages — and so is whether the new Monetary Policy Report shows inflation coming back to 2% with or without further rises.

After that comes the Fed on 9 December, the first meeting where Wednesday’s two dissenters can win, and the ECB on 17 December with new projections.

I do not know whether December brings a Fed rise. Nobody does, and the chair himself would not say. What I do know is where to look when the statement lands: the probability before, the words that changed, the names at the bottom.

The biggest gap in the market right now is between a Fed that is pausing and a Bank of Japan that is still climbing, and that gap lives on the yen pairs. For what each move there is worth in money, see what one pip is worth on USD/JPY.

Sources: Federal Reserve, FOMC statements of 29 July, 16 September and 28 October 2026; European Central Bank, monetary policy decision of 29 October 2026; Bank of England, Monetary Policy Summary and minutes, September 2026 and the earlier 2026 minutes for the vote counts; Bank of Japan, monetary policy decisions, 18 September and 30 October 2026. Market probabilities are as reported on 28–30 October 2026 and change every day.