Succinct Summation of the Week's Events

Succinct Summation of the Week’s Events:

Positives,

1)Personal income in January surprised to the upside with a .9% m/o/m pop, more than double the estimate of up .4%. Private sector wages and salaries though were up by a more modest .4% m/o/m which was the same amount seen in December. The reason for the headline beat was a 39% m/o/m spike in farmers income and a 1.8% rise in transfer payments. Combining the rise in income and the drop in spending, the savings rate rose to 4.6% from 3.5%.

2)Core durable goods orders in January rose .8% m/o/m, well above the estimate of up .3%, partly offset by a 2 tenths downward revision to December.

3)New home sales in January totaled 657k, below the estimate of 680k but December was revised up by 36k to 734k. Smoothing it out, the 3 month average is now 690k vs the 6 month average of 685k and which compares with the 2024 average of 686k. For further perspective, the average in 2019 was 685k. Months’ supply rose to 9.0 from 8.0 in December and 8.6 in November. The absolute number of new homes on the market now total 495k, the most since December 2007. The median home price, highly influenced by mix, rose 3.7% y/o/y and 7.5% m/o/m.

4)The Treasury 3s, 5s and 7 yr auctions went well this week.

5)The February Richmond manufacturing index rose to +6 from -4.

6)From ZipRecruiter: "Despite the protracted labor market downturn, we enter 2025 with cautious optimism from both internal and external indicators. The NFIB's Small Business Optimism index in December posted its highest reading since October of 2018, which can be a leading indicator for employer hiring plans. There are other encouraging underlying signs internally, such as an uptick in employer account reactivations."

7)From Nvidia: "Post training and model customization are fueling demand for Nvidia infrastructure and software as developers and enterprises leverage techniques such as fine-tuning, reinforcement learning, and distillation to tailor models for domain specific use cases...The scale of post training and model customization is massive and can collectively demand orders of magnitude more compute than pre-training. Our inference demand is accelerating, driven by test time scaling and new reasoning models like Open AI's o3, DeepSeek-R1, and Grok 3."