3 Takes From Our AI Model on Powell’s Press Conference

We have been building a series of AI models that interpret central bank speeches. Unlike many existing models, ours use the full capabilities of ChatGPT (rather than older BERT models), and we go beyond analysing just sentences. It is still a work-in-progress, but here are three takes on yesterday’s Powell presser:

1. Powell’s remarks and Q&A were more dovish yesterday than in the January meeting. One version of our model scores it as becoming 0.4 points more dovish, while our more refined model scores it as becoming 1 point more dovish. (We have a 20-point scale).
2. Powell struggled most answering a question on financial conditions yesterday. In January, he struggled most answering questions on Fed confidence on inflation.
3. The newest question was on CBDC. Powell was not asked about it in January but was asked yesterday. Powell fended this off as a decision for Congress. But it will likely show how CBDC is a political question and come back again in future press conferences.

I’ll keep you posted on the development of our models!

Bilal

.

Bilal Hafeez is the CEO and Editor of Macro Hive. He spent over twenty years doing research at big banks – JPMorgan, Deutsche Bank, and Nomura, where he had various “Global Head” roles and did FX, rates and cross-markets research.

.

(The commentary contained in the above article does not constitute an offer or a solicitation, or a recommendation to implement or liquidate an investment or to carry out any other transaction. It should not be used as a basis for any investment decision or other decision. Any investment decision should be based on appropriate professional advice specific to your needs.)