Last week, the leasehold and reform bill became law in the UK. This is not the end of leasehold reform. It is not even the beginning of the end. But it is, perhaps, the end of the beginning. This passage of the bill happened against the odds. It was crawling through parliament and when Rishi Sunak, somewhat unexpectedly, announced the general election on July the 4th, campaigners for leasehold reform—a group which yours truly have been loosely working with for a while—thought the bill would be lost. If you want to understand what happens to outstanding bills in the brief final sessions of parliament before prorogation, you need to read to up on something called wash-up. Put simply, it’s the period where outstanding bills are either rammed through as is, or kicked into the abyss never to be seen again. It is a borderline insane policy process, which breaks all the rules of legislation, simply to get the order book emptied as quickly as possible. The leasehold and reform bill made it, just, though without key additions such as a cap on ground rent or a ban on forfeiture. This is bitterly disappointing, especially in case of the former given that an agreement to cap ground rents to £250 pa was virtually agreed by the department and HMT, or so we’re told. But I guess we live to fight another day rather than having to start over.
Read MoreThe BIS has a Bulletin out on the usefulness of AI and large language models. They’re not terribly impressed.
Read MoreWhen posed with a logical puzzle that demands reasoning about the knowledge of others and about counterfactuals, large language models (LLMs) display a distinctive and revealing pattern of failure.
The LLM performs flawlessly when presented with the original wording of the puzzle available on the internet but performs poorly when incidental details are changed, suggestive of a lack of true understanding of the underlying logic.
Our findings do not detract from the considerable progress in central bank applications of machine learning to data management, macro analysis and regulation/supervision. They do, however, suggest that caution should be exercised in deploying LLMs in contexts that demand rigorous reasoning in economic analysis.